docs: update README.md to clarify file generation process and avoid temporary files

Co-authored-by: Copilot <copilot@github.com>

Author: Arnaud Riess

src/extensions/score_metamodel/README.md

@@ -99,13 +99,13 @@ Schema column: **yes** = implemented, **feasible** = could be added, **--** = no
 | Mandatory link ID regex | feasible | yes | Can add `items.pattern` in local; TODO in code |
 | Optional link target type | feasible | yes (info) | Split into separate schema with `severity: "info"` |
 | Optional link ID regex | feasible | yes (info) | Same split-severity approach |
-| Mixed regex+plain link type | -- | yes | `ValidateSchemaType` has no `anyOf`/`oneOf` |
+| Mixed regex+plain link type | to be analyzed | yes | `ValidateSchemaType` has no `anyOf`/`oneOf` |
 | ID structure (parts count) | feasible | yes | Per-type pattern from `parts` field; cannot check file-path part |
 | Prohibited words | feasible | yes | Negative lookahead regex on `title`; less precise than Python |
-| Graph constraints | -- | yes | Cross-need traversals beyond JSON Schema |
-| Undefined extra options | -- | yes | `unevaluatedProperties` would reject sphinx-needs internal fields |
+| Graph constraints | to be analyzed | yes | Cross-need traversals beyond JSON Schema |
+| Undefined extra options | to be analyzed | yes | `unevaluatedProperties` would reject sphinx-needs internal fields |
 | ID length limit (45 chars) | feasible | yes | Can add `maxLength` to per-type ID pattern in schema |
-| Validity period consistency (`valid_from` < `valid_until`) | -- | yes | Cross-field comparison beyond JSON Schema |
+| Validity period consistency (`valid_from` < `valid_until`) | to be analyzed | yes | Cross-field comparison beyond JSON Schema |
 
 #### Rule explanations
 

src/extensions/score_sync_toml/README.md

@@ -70,9 +70,11 @@ directives as needs** and the index is empty.
    `metamodel.yaml` before `score_sync_toml` runs.
 2. Generating the `[[needs.types]]` TOML entries in memory with only the fields
    ubCode requires: `directive`, `title`, `prefix`, and optionally `color` / `style`.
-3. Writing the result (combined with fields and links) into a temporary file and
-   appending it to `needscfg_merge_toml_files` so it is merged into the
-   final `ubproject.toml`.
+3. Writing the result (combined with fields and links) to
+   `needs_metamodel_generated.toml` in `confdir` and appending it to
+   `needscfg_merge_toml_files` so it is merged into the final `ubproject.toml`.
+   A deterministic path is used so the file is always overwritten and no
+   temporary files accumulate.
 
 ### 5. Relative path handling
 
@@ -106,7 +108,7 @@ score_metamodel.setup()  ──► app.config.needs_types (Python objects)
                             _generate_needs_links_toml()
                                          │
                                          ▼
-                            <temp file>.toml ──────────────────┐
+                     needs_metamodel_generated.toml ───────────┐
                                                                │
 shared.toml  ──────────────────────────────────────────────────┤  merge via
                                                                │  needs_config_writer
@@ -123,11 +125,12 @@ app.config (all other needs_* settings)  ─────────────
 
 ## Generated files
 
-Both files are written to `confdir` (the `docs/` directory) on every Sphinx build
-and are committed to the repository so that ubCode works without needing to run a
-build first.
+All three files are written to `confdir` (the `docs/` directory) on every Sphinx
+build and are committed to the repository so that ubCode works without needing to
+run a build first.
 
 | File | Generated by | Used by |
 |---|---|---|
 | `ubproject.toml` | `needs_config_writer` + merges above | ubCode language server |
 | `schemas.json` | `score_metamodel.sn_schemas.write_sn_schemas` | ubCode + sphinx-needs schema validation |
+| `needs_metamodel_generated.toml` | `score_sync_toml` (`_generate_needs_types_toml` etc.) | Intermediate merge input for `needs_config_writer` |