Bump version to 2.7.6

[igerber]

Summary

• Bump version from 2.7.5 to 2.7.6 across all version locations
• Add CHANGELOG entry covering PRs Add survey Phase 6: replicate weights, DEFF diagnostics, subpopulation analysis #238 and Add AI practitioner guardrails (Baker et al. 2025) #239

Methodology references (required if estimator / math changes)

• N/A — no methodology changes

Validation

• Tests added/updated: No test changes (version bump only)
• Backtest / simulation / notebook evidence: N/A

Security / privacy

• Confirm no secrets/PII in this PR: Yes

Files changed

• diff_diff/__init__.py — __version__ = "2.7.6"
• pyproject.toml — version = "2.7.6"
• rust/Cargo.toml — version = "2.7.6"
• docs/llms-full.txt — Version: 2.7.6
• CHANGELOG.md — Added [2.7.6] entry + comparison link

---

Generated with Claude Code

[github-actions]

Overall Assessment

✅ Looks good

Executive Summary

• The diff is limited to release metadata and release notes in diff_diff/init.py:197, pyproject.toml:7, rust/Cargo.toml:3, docs/llms-full.txt:5, and CHANGELOG.md:10.
• No estimator implementation, weighting logic, variance/SE path, identification assumption check, or default behavior changed in this PR.
• The methodology-heavy 2.7.6 changelog bullets are already reflected in the Methodology Registry, including Callaway-Sant’Anna survey SEs, Phase 6 survey bootstrap, DEFF/subpopulation support, and the practitioner guide at REGISTRY.md:419, REGISTRY.md:501, REGISTRY.md:683, REGISTRY.md:2067, REGISTRY.md:2086, and REGISTRY.md:2116.
• Version surfaces are internally consistent across Python, Rust, and the AI-facing full-text doc.
• I did not find any unmitigated P0 or P1 issues in scope.

Methodology

• Severity: P3 informational. Impact: no methodology defect in scope. This PR only updates version strings and the release note, while the substantive methodology items called out in CHANGELOG.md:13 and CHANGELOG.md:18 are already documented in the registry at REGISTRY.md:419, REGISTRY.md:501, REGISTRY.md:683, REGISTRY.md:2067, REGISTRY.md:2086, and REGISTRY.md:2116. Concrete fix: none.

Code Quality

• Severity: P3 informational. Impact: no code-quality issue found in the changed files. The bumped version is consistent in diff_diff/init.py:197, pyproject.toml:7, rust/Cargo.toml:3, and docs/llms-full.txt:5. Concrete fix: none.

Performance

• Severity: P3 informational. Impact: none. The PR does not touch estimator runtime paths, inference routines, or build-hot code. Concrete fix: none.

Maintainability

• Severity: P3 informational. Impact: none. Documentation versioning already derives from diff_diff.__version__ in docs/conf.py:19, so this bump remains aligned with the docs build. Concrete fix: none.

Tech Debt

• Severity: P3 informational. Impact: none. This PR does not introduce new deferred work and does not rely on TODO-based mitigation; existing review-tracked debt remains unchanged under TODO.md:44. Concrete fix: none.

Security

• Severity: P3 informational. Impact: none. The diff changes release metadata and changelog text only; I found no new dependency, secret, or executable-surface risk. Concrete fix: none.

Documentation/Tests

• Severity: P3 informational. Impact: none. The new release note is consistent with existing practitioner-facing artifacts in diff_diff/practitioner.py:49, docs/practitioner-guide-evaluation.md:8, docs/llms.txt:27, and docs/llms-practitioner.txt:15. No new tests are required for a metadata-only release bump. Concrete fix: none.