Release v2.9.1

[igerber]

Summary

• Bump version to 2.9.1 across all version locations
• Add CHANGELOG entry covering: survey theory document, research-grade DGP (8 params), R validation expansion (4 estimators), tutorial rewrite, Wooldridge carousel, documentation refresh and fixes

Methodology references

• N/A — version bump only, no methodology changes in this PR

Validation

• No test changes — release bookkeeping only

Security / privacy

• Confirm no secrets/PII in this PR: Yes

---

Generated with Claude Code

[github-actions]

Overall Assessment

✅ Looks good

No unmitigated P0/P1 issues found in the diff.

Executive Summary

• The diff is limited to release metadata and changelog text in CHANGELOG.md:L10, diff_diff/init.py:L217, pyproject.toml:L7, rust/Cargo.toml:L3, and docs/llms-full.txt:L5.
• No estimator, weighting, variance/SE, identification-assumption, or default-behavior code changed, so I found no new methodology-registry or paper-adherence issues.
• Version strings are synchronized across the touched Python, Rust, and documentation surfaces.
• Spot checks of the new release notes matched current repo state: survey-theory.md:L1 exists, the survey DGP parameters listed in the changelog exist in prep_dgp.py:L1184, and the llms-full.txt fixes referenced in the notes are present at docs/llms-full.txt:L1550 and docs/llms-full.txt:L1568.
• One existing non-blocking gap remains relevant: .txt AI-guide docs are not smoke-tested in CI, but that limitation is already tracked in TODO.md:L92.

Methodology

• No findings. I checked docs/methodology/REGISTRY.md, and nothing in this diff changes estimator logic, math, weighting, variance/SE, assumption checks, or defaults.

Code Quality

• No findings. The version bump is internally consistent across diff_diff/init.py:L217, pyproject.toml:L7, rust/Cargo.toml:L3, and docs/llms-full.txt:L5.

Performance

• No findings. No runtime code paths or computational kernels changed.

Maintainability

• No findings. The release bookkeeping adds the compare link at CHANGELOG.md:L1176, and the visible version surfaces touched by this PR are synchronized.

Tech Debt

• No findings. TODO.md is unchanged by this diff, and the PR does not introduce new deferred correctness work.

Security

• No findings. The changed files add no secrets, credentials, dependency changes, or new input-handling surface.

Documentation/Tests

• P3 Impact: docs/llms-full.txt:L5 is in a documentation surface that current smoke tests do not validate; drift in .txt AI-guide examples can still escape CI. This gap is already tracked in TODO.md:L92, so it is informational only and does not affect approval. Concrete fix: extend tests/test_doc_snippets.py to cover .txt AI-guide files when you want automated validation for this surface.