Validate baseline.strategy; align docs with implemented metrics

[quantecon-services]

Phase 0 of the work plan in #8: the two defects found by the 2026-08-25 audit, plus the lint fixes needed to keep CI green under current ruff.

What changed

F1 — baseline.strategy was accepted but never read. load_config now validates it: unknown values are rejected, and the planned but unimplemented state-file strategy raises a clear "planned but not implemented" error, mirroring the existing guard for prose.strategy: source-diff. Previously baseline: {strategy: state-file} — which #1 directs the operator to try on programming.zh-cn — was silently accepted and scanned with script-jump, producing plausible-but-wrong baselines. Added test_baseline_strategy_validated covering both the unimplemented value and a typo.

F2 — docs claimed character-level churn that does not exist. README said churn "is also reported in changed characters" and docs/method.md said it is "available in changed characters from the pairs"; no artefact carries a character count. Both now say character-level churn is planned and point at #4 (no closing keyword — this PR does not implement it).

Lint. ruff 0.15 (CI installs ruff unpinned, >=0.6) now enforces E701/E702/E741, which flagged 7 pre-existing errors in git.py, scan.py and tests/test_units.py; fixed mechanically (renamed ambiguous l variables, split multi-statement lines) so this PR's CI — and the next push to main — stays green.

Also adds CHANGELOG entries for both fixes.

Validation

ruff check src tests clean; pytest -q 7 passed, 1 skipped (the regression test needs the pinned lecture-intro.zh-cn checkout and runs in CI). No behaviour change for any existing config: all shipped configs use script-jump, and the regression-compared quantities are untouched.

🤖 Generated with Claude Code

https://claude.ai/code/session_01MhveTMPqgH3AhmJui3h41m

---

Generated by Claude Code

[Copilot]

Pull request overview

This PR lands Phase 0 of the work plan in issue #8: two audit defect fixes plus the lint fixes needed to keep CI green under current ruff. It fits into the deterministic scan pipeline by closing a silent-misconfiguration gap in load_config and correcting documentation that overstated the implemented metrics, without altering any scanning behaviour (all shipped configs use script-jump).

Changes:

• F1: load_config now validates baseline.strategy, rejecting unknown values and raising a clear "planned but not implemented" error for state-file, mirroring the existing prose.strategy guard.
• F2: README and docs/method.md no longer claim character-level churn is reported; they now mark it as planned and point to #4.
• Mechanical ruff E701/E702/E741 fixes in git.py, scan.py, and tests/test_units.py, plus a new validation test and CHANGELOG entries.

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
src/textstrata/config.py	Adds baseline.strategy validation and reuses the validated instance in Config.
tests/test_units.py	Adds test_baseline_strategy_validated; splits multi-statement lines to satisfy ruff.
src/textstrata/scan.py	Renames ambiguous l loop variable to ln (E741).
src/textstrata/git.py	Renames ambiguous l loop variable to ln (E741).
README.md	Marks character-level churn as planned, linking #4.
docs/method.md	Aligns the known-limits note with implemented metrics, linking #4.
CHANGELOG.md	Records the F1 validation and F2 doc-alignment fixes.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.