feat(ui): MVP UX polish from the heuristic evaluation

CHANGELOG.md

@@ -217,6 +217,48 @@ _Feature module implementation (iam, billing, workspace, discovery) in progress.
 
 ### Changed
 
+- **UX — MVP usability polish** (`feature/mvp-ux-polish`): a refinement pass on the MVP surfaces driven by
+  a heuristic evaluation with 6 users (Nielsen, impeccable critique: live session 20 → 26/40, stories
+  20 → 24/40), keeping the brand, behaviour and copy.
+  - *Live session*: the AI suggestion queue docks to the feed column as a framed **review tray** ("AI
+    suggestions to review", counter with prev/next, minimize) instead of floating over the session bar,
+    header actions and side panel; from `sm` it stops short of the feed bottom so the latest lines stay
+    visible. ←/→ browse it, Esc minimizes it, focus returns to it after a decision, and focusing a feed
+    control it covers minimizes it (WCAG 2.4.11). A status line at the live edge says what the AI is doing — listening
+    (with the last suggestion's age), paused, **processing the final stories after Stop**, or failed with
+    the backend reason. The transcript recedes into neutral bubbles with visible speaker/time
+    ("Participante n" in Spanish); human decisions become compact "accepted/resolved by the analyst"
+    rows; accepting confirms that the story landed in the backlog as a draft.
+  - *AI vs human provenance*: new `ai` / `verified` / `pending` tokens. Everything the AI proposed reads
+    violet (suggestion cards, generated stories, "IA" origin chip); human-validated content reads emerald;
+    drafts awaiting review read amber. Priority gets its own glyph scale (critical red, high orange,
+    medium/low neutral), so red is no longer brand, AI, danger and priority at once.
+  - *Readable stories*: `hlmInput` textareas grow with their content (they were clipped at a fixed
+    height); stories read as their sentence and criteria as **Gherkin** steps with a keyword gutter in the
+    suggestion card, side panel, story detail and create page. The story detail leads with the title,
+    review status (with what it means) and origin, links an AI story to its capture session, and flags
+    unsaved story and criterion edits.
+  - *Backlog order*: sortable headers with a direction arrow and `aria-sort`, a visible "sorted by"
+    caption, priority re-ordered by meaning on the page (the API sorts the enum alphabetically), origin
+    and status chips (draft as an amber dashed outline), a page-scrolling chip-row layout below `sm`,
+    unambiguous dates, an empty state with next steps and retry on load errors. Story forms flag empty
+    required fields inline.
+  - *Audio source*: the picker is a labelled segmented control (In person / Virtual meeting) next to a
+    record button with a record dot and label; the virtual-meeting guidance is a calm two-step note
+    (share the meeting tab with its audio; wear headphones) and the missing-audio notice offers "Share
+    again" in place.
+  - *Project forms*: only the name is required and both forms say so; the optional technical profile is
+    one group (business context + tech stack, example placeholders, Enter-to-add hint). Settings go from
+    eight one-field cards to General and Technical profile, and the logo card is titled for the project.
+  - *Accessibility*: the brand red fill is `#dc2626` so white labels pass AA (4.83:1), red text in dark
+    uses a lighter step, muted text and the default focus outline meet AA, filter inputs show focus,
+    story titles are keyboard links, panel tabs/sort chips expose their state, the mobile session-bar
+    status stays announced, and new suggestions are announced politely. All measured chips are ≥ 4.5:1
+    in both themes.
+  - Report screenshots (light, 1440×900, Spanish demo data) in `docs/screenshots/mvp/`. New
+    `discovery.ai.*`, `discovery.queue.*`, `discovery.suggestion.*`, `stories.statusHint.*`,
+    `stories.origin.*`, `storyForm.*`, `projectCreate.*`, `projectSettings.*` and `common.optional/required`
+    keys (EN + ES); `discovery.source.hint` is replaced by `hintShare` / `hintHeadphones`.
 - **Core — centralized backend error handling** (`feature/frontend-error-handling`): a shared
   `messageForError` helper resolves backend errors by their machine-readable `code` against a single
   top-level `errors.<CODE>` i18n block (network / per-status / generic fallback chain). Extended from the

public/i18n/en.json

@@ -7,7 +7,9 @@
     "create": "Create",
     "loading": "Loading",
     "back": "Back",
-    "remove": "Remove"
+    "remove": "Remove",
+    "optional": "Optional",
+    "required": "Required"
   },
   "toast": {
     "dismiss": "Dismiss",
@@ -374,8 +376,14 @@
   "projectCreate": {
     "title": "New project",
     "subtitle": "Give it a name. You can add the tech profile later — it helps the AI but is optional.",
-    "advanced": "Advanced options",
-    "advancedHint": "Optional context the AI uses to tailor its suggestions."
+    "advanced": "Tech profile",
+    "advancedHint": "Optional context the AI uses to tailor its suggestions.",
+    "namePlaceholder": "e.g. Patient portal",
+    "descriptionPlaceholder": "What problem does it solve, and for whom?",
+    "contextGroup": "Business context",
+    "stackGroup": "Tech stack",
+    "chipHint": "Type a value and press Enter to add it.",
+    "example": "e.g. {{value}}"
   },
   "orgPicker": {
     "title": "Choose an organization",
@@ -670,7 +678,10 @@
     "saved": "Changes saved.",
     "save": "Save changes",
     "errorValidation": "Check the data and try again.",
-    "errorGeneric": "Could not save the changes."
+    "errorGeneric": "Could not save the changes.",
+    "logo": "Project logo",
+    "general": "General",
+    "techProfile": "Tech profile"
   },
   "projectRoles": {
     "title": "Roles",
@@ -882,7 +893,23 @@
       "REJECTED": "Rejected",
       "MERGED": "Merged",
       "EXPORTED": "Exported"
-    }
+    },
+    "statusHint": {
+      "DRAFT": "Draft: not approved yet; you can still edit it before approving.",
+      "APPROVED": "Approved: validated by the team.",
+      "REJECTED": "Rejected: discarded by the team.",
+      "MERGED": "Merged: folded into another story.",
+      "EXPORTED": "Exported: pushed to an external tracker (e.g. Jira)."
+    },
+    "origin": {
+      "ai": "AI",
+      "manual": "Manual",
+      "aiHint": "Generated by the AI in a capture session",
+      "manualHint": "Created manually"
+    },
+    "colOrigin": "Origin",
+    "sortedBy": "Sorted by: {{sort}}",
+    "emptyRecord": "Record a session"
   },
   "storyForm": {
     "back": "Back to stories",
@@ -918,7 +945,16 @@
     "criteriaSaved": "Criterion saved.",
     "criteriaDeleted": "Criterion removed.",
     "criteriaIncomplete": "Fill in Given, When and Then for each criterion.",
-    "criteriaPartial": "Story created, but {{count}} criterion(s) could not be saved. Add them here."
+    "criteriaPartial": "Story created, but {{count}} criterion(s) could not be saved. Add them here.",
+    "storySection": "User story",
+    "leadRole": "As a…",
+    "leadAction": "I want…",
+    "leadBenefit": "so that…",
+    "viewSession": "View source session",
+    "unsaved": "Unsaved changes",
+    "criteriaNew": "New · not saved",
+    "criteriaHint": "Write each criterion as a Gherkin scenario: Given (context), When (action), Then (expected outcome).",
+    "required": "Fill in this field."
   },
   "glossaryPage": {
     "title": "Glossary",
@@ -1044,7 +1080,7 @@
     "sessionSeparator": "Session",
     "sessionStories": "{{count}} stories",
     "sessionDefaultTitle": "Discovery session — {{date}}",
-    "generatedStory": "Generated story",
+    "generatedStory": "Generated by AI",
     "speaker": "Speaker {{n}}",
     "goToStory": "Go to story",
     "time": {
@@ -1062,7 +1098,11 @@
     },
     "decision": {
       "accepted": "Accepted",
-      "dismissed": "Dismissed"
+      "dismissed": "Dismissed",
+      "acceptedBy": "Accepted by the analyst",
+      "resolvedBy": "Resolved by the analyst",
+      "acceptedToast": "Suggestion accepted: the story is in the backlog as a draft.",
+      "resolvedToast": "Question marked as resolved."
     },
     "live": {
       "banner": "Live session in progress — {{language}}"
@@ -1074,7 +1114,8 @@
     "composer": {
       "placeholder": "Message discovery…",
       "comingSoon": "Text to pipeline is coming soon.",
-      "record": "Start recording"
+      "record": "Start recording",
+      "recordShort": "Record"
     },
     "language": {
       "label": "Meeting language",
@@ -1087,7 +1128,10 @@
       "mic": "In person (microphone)",
       "meeting": "Virtual meeting (microphone + meeting audio)",
       "unsupported": "Virtual meeting: available in desktop Chrome or Edge.",
-      "hint": "When sharing, pick the meeting tab and tick \"Share tab audio\". Use headphones to avoid a duplicated transcript."
+      "micShort": "In person",
+      "meetingShort": "Virtual meeting",
+      "hintShare": "When you press Record, pick the meeting tab and turn on “Share tab audio”.",
+      "hintHeadphones": "Wear headphones so your microphone doesn't pick up the meeting again and duplicate the transcript."
     },
     "bar": {
       "recording": "Recording",
@@ -1097,7 +1141,8 @@
       "stop": "Stop",
       "meetingAudio": "Recording microphone + meeting audio",
       "meetingAudioLost": "Meeting audio is not being captured. Share the meeting tab again.",
-      "reshare": "Share again"
+      "reshare": "Share again",
+      "meetingAudioShort": "Meeting audio"
     },
     "rec": {
       "micDenied": "Microphone permission denied. Enable it in your browser settings.",
@@ -1113,13 +1158,17 @@
     },
     "queue": {
       "pendingBadge": "{{count}} suggestions pending",
-      "pendingBadgeLabel": "pending suggestions",
+      "pendingBadgeLabel": "suggestions to review",
       "counter": "{{n}} of {{m}}",
       "prev": "Previous suggestion",
       "next": "Next suggestion",
       "collapse": "Collapse",
       "minimize": "Minimize",
-      "dragHint": "Drag to browse suggestions"
+      "dragHint": "Drag to browse suggestions",
+      "title": "AI suggestions to review",
+      "announce": "{{count}} AI suggestions to review",
+      "expand": "Show the suggestions to review",
+      "titleShort": "AI · to review"
     },
     "suggestion": {
       "type": {
@@ -1151,7 +1200,7 @@
       "saveAndAccept": "Save & accept",
       "dismiss": "Dismiss",
       "edit": "Edit",
-      "cancelEdit": "Cancel",
+      "cancelEdit": "Cancel editing",
       "done": "Done",
       "titleField": "Title",
       "scenario": "Scenario",
@@ -1165,7 +1214,16 @@
       "removeCriterion": "Remove criterion",
       "given": "Given",
       "when": "when",
-      "then": "then"
+      "then": "then",
+      "aiLabel": "AI suggestion",
+      "aiHint": "Proposed by the AI from the conversation. Awaiting your review.",
+      "questionHint": "Resolving it only marks it as addressed; the backlog doesn't change.",
+      "changedHint": "Highlighted: what the AI proposes to change",
+      "criteriaCount": "Acceptance criteria ({{count}})",
+      "roleField": "Role · As a…",
+      "actionField": "Action · I want…",
+      "benefitField": "Benefit · so that…",
+      "scenarioPlaceholder": "Optional"
     },
     "panel": {
       "stories": "Stories",
@@ -1227,6 +1285,19 @@
       "body": "Leave this page anyway? The session keeps recording in the background and stays active.",
       "stay": "Stay",
       "leave": "Leave anyway"
+    },
+    "retry": "Retry",
+    "pendingReview": "Awaiting review",
+    "ai": {
+      "listening": "The AI is listening to the conversation",
+      "listeningHint": "It proposes stories, edge cases and questions as the meeting goes on; they appear above for your review.",
+      "lastSuggestion": "Last suggestion {{time}} · review it in the tray above.",
+      "paused": "Recording paused",
+      "pausedHint": "The AI resumes its analysis when you resume recording.",
+      "processing": "Processing the session",
+      "processingHint": "The AI is generating the final stories. This can take a few minutes.",
+      "failed": "The session could not be processed",
+      "failedHint": "Check the session history or try again later."
     }
   },
   "integrations": {