fix: migrate generic UI controls to Radix

[HomenShum]

What changed

• keeps the compact chat answer typography fix and continuation behavior already in this PR
• migrates generic dialogs, sheets, popovers, hover cards, menus, command palettes, tabs, toggle groups, switches, checkboxes, accordions, and toast delivery to shared shadcn/Radix primitives
• removes bespoke focus-trap, bottom-sheet, keyboard-navigation, portal-positioning, and duplicate toast infrastructure
• preserves native browser form controls, route navigation, editor-decoration portals, and product-specific virtualized menus
• fixes the browser-discovered report action-menu runtime import and duplicate live-artifact rail keys
• adds source guards and focused behavior tests to prevent handwritten interaction plumbing from returning

Why

The app had many visually similar controls with separate Escape handling, focus behavior, outside-click logic, portal geometry, ARIA state, and toast stores. Centralizing these behaviors reduces UI clutter and makes keyboard, screen-reader, collision, and mobile behavior consistent without changing runtime-backed product actions.

Verification

• npx tsc --noEmit --pretty false
• 31 final migration-focused tests passed
• 297 broader FastAgent, documents, entity, layout, and design-system regressions passed
• npm run lint:design (0 high-severity violations; existing medium/low baseline advisories)
• post-fix npm run build
• in-app browser: command palette keyboard/focus return, reports ToggleGroups and action menu, mobile chat at 390x844, zero horizontal overflow
• clean production-data reports rerun: zero alerts, zero console errors, five Radix rail radios, six menu items

Intentional boundary

Native <select> and simple <details> elements remain browser primitives. Route navigation remains navigation. Editor slash/mention/entity disambiguation and editor-decoration portals remain product-specific composites rather than being incorrectly forced into generic Tabs/Menu abstractions.

[github-actions]

Tier B regression failed against preview

Preview URL: (preview URL unresolved)
Run: 29540394656

Likely a kit-parity DOM regression. Check the Playwright report artifact attached to the run for the failing selector + screenshot.

This check would have caught all 4 A9 sessionRows iterations as a single PR — that's why it exists.

[github-actions]

Tier B regression failed against preview

Preview URL: (preview URL unresolved)
Run: 29540962422

Likely a kit-parity DOM regression. Check the Playwright report artifact attached to the run for the failing selector + screenshot.

This check would have caught all 4 A9 sessionRows iterations as a single PR — that's why it exists.

[github-actions]

Tier B regression failed against preview

Preview URL: (preview URL unresolved)
Run: 29541555405

Likely a kit-parity DOM regression. Check the Playwright report artifact attached to the run for the failing selector + screenshot.

This check would have caught all 4 A9 sessionRows iterations as a single PR — that's why it exists.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
nodebench-ai	Ready	Preview, Comment	Jul 17, 2026 12:22am

[github-actions]

✅ Dogfood Visual QA Gate: PASSED

Check	Status
Screenshots	23 captured (pass)
Walkthrough	9 chapters (pass)
Key Frames	9 extracted (pass)
Scribe Steps	8 how-to steps (pass)
Build	success

Artifacts

Download the dogfood-evidence-e181bd9 artifact from the Actions tab for full screenshots, frames, and walkthrough video.

---

Generated by Dogfood QA Gate

[github-actions]

Demo: walkthrough of the surfaces this PR changed is available as a workflow artifact (pr-demo-559) at https://github.com/HomenShum/NodeBenchAI/actions/runs/29544826478