fix(forecast,docs): address CodeRabbit feedback on PR #266 (3 quick wins) (#267)

[w7-mgfcode]

Closes #267.

Three quick-win CodeRabbit comments from PR #266 (the v0.2.17 dev → main release prep). All flagged in code/docs that landed in the release; tackled here as a follow-up so the release window stayed clean.

Commits (one per fix, individually revertable)

SHA	Type	Summary
79db58f	fix(forecast)	Validate bundle_path_str type in feature-metadata loader
1ea196f	docs(docs)	Fix PRP-31 endpoint count in scope statement
655007a	docs(docs)	Align PRP-31 missing-bundle status code to 422

What each fix does

1. fix(forecast) — CodeRabbit #3270767550

app/features/forecasting/service.py:809-812 only checked truthiness on job.result["model_path"] (an untyped JSONB column). A non-string truthy value (int/float/bool/list/dict) would bypass the guard and surface later as an unhandled TypeError inside load_model_bundle. Strengthened the guard to isinstance(bundle_path_str, str) and bundle_path_str.strip() and added a parametrized regression test (7 cases) covering int, float, bool, list, dict, empty string, and whitespace.

2. docs(docs) — CodeRabbit #3270767557

PRP-31 lines 17-18 said "one new backend endpoint" but the PRP defines two (/forecasting/runs/{run_id}/feature-metadata and /forecasting/jobs/{job_id}/feature-metadata). Rephrased the scope statement to name both paths inline so it matches the detailed spec at lines 97-99, 149, and 567.

3. docs(docs) — CodeRabbit #3270767563

PRP-31 line 1210 listed 500 — bundle file missing on disk but the rest of the PRP and the implementation (forecasting/service.py:832-839) raise UnprocessableEntityError → 422 for the deleted-artifact case. Folded the "missing on disk" bullet into the 422 line so the documented contract matches both the code and test_routes_feature_metadata.py::test_get_run_metadata_returns_422_for_deleted_artifact.

Out of scope

CodeRabbit's fourth comment (#3270767554) flagged the registry↔forecasting ModelFamily slice cycle. That's a heavy-lift refactor needing its own PRP and is tracked separately in #268.

Test plan

• uv run ruff check . — clean
• uv run ruff format --check . — clean (299 files)
• uv run mypy app/features/forecasting/ — Success: no issues found in 21 source files
• uv run pyright app/features/forecasting/service.py app/features/forecasting/tests/test_routes_feature_metadata.py — 0 errors / 0 warnings
• uv run pytest -v -m "not integration" app/features/forecasting/ — 187 passed

[sourcery-ai]

Sorry @w7-mgfcode, you have reached your weekly rate limit of 500000 diff characters.

Please try again later or upgrade to continue using Sourcery

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: ceb23719-4150-4975-a4a9-4e2cd7ae4821

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch fix/coderabbit-pr266-quickwins

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}