Release 3.6.1

[igerber]

Summary

• Bump version 3.6.0 → 3.6.1 across diff_diff/__init__.py, pyproject.toml, rust/Cargo.toml, diff_diff/guides/llms-full.txt, and CITATION.cff.
• Promote the CHANGELOG.md [Unreleased] section to [3.6.1] - 2026-07-01 (fresh empty [Unreleased] retained on top) and add the v3.6.0...v3.6.1 comparison link.

This is a release-cut PR: no source/behavior changes of its own — it tags the work already merged to main since v3.6.0. Release highlights:

• LP-DiD non-absorbing (reversible) treatment — entry-effect estimands (Dube, Girardi, Jordà & Taylor 2025), R-parity validated (Add LPDiD non-absorbing treatment (entry-effect estimands), Phase C1 #584, Add LPDiD non-absorbing R-parity (independent feols Eq.12/13), Phase C2 #587).
• LP-DiD complex-survey-design support — new survey_design= argument, R-parity validated vs survey::svyglm (Add LPDiD complex-survey-design support (Phase D1) #590, Add LPDiD complex-survey R-parity validation vs survey::svyglm (Phase D2) #591).
• TROP non-absorbing (on/off) treatment — opt-in, local method (Athey, Imbens, Qu & Viviano 2025) (Add TROP non-absorbing (on/off) treatment support (opt-in, local method) #592).
• Weighted multiple absorbed fixed effects — absorb=[a, b, ...] via iterative alternating-projection demeaning (feat(estimators): iterative alternating-projection demeaning for N-way absorbed FE #586).
• CallawaySantAnna — non-estimable (g,t) cells now materialized as NaN (feat(staggered): materialize non-estimable (g,t) cells as NaN in CallawaySantAnna #582).
• Fix: BusinessReport appendix render failures now surfaced (fix: surface BusinessReport appendix render failures #574).
• Docs/validation: survey zero-weight-PSU SE-invariance item waived (Lumley full-design convention) (docs(survey): waive zero-weight-PSU SE-invariance item; lock Lumley full-design convention #589); SciPy lower-bound doc alignment (docs: align SciPy lower bound with package metadata #572).

Methodology references (required if estimator / math changes)

• Method name(s): N/A for this PR — version/changelog only. The estimators/fixes it releases were reviewed in their own merged PRs, each citing its source in docs/methodology/REGISTRY.md.
• Paper / source link(s): N/A (no methodology code in this diff)
• Any intentional deviations from the source (and why): None

Validation

• Tests added/updated: None (no source changes in this PR; the released work shipped with its own tests + R-parity harnesses).
• Backtest / simulation / notebook evidence (if applicable): N/A
• Note: recommend running the full-matrix wheel dry-run (release-build-check.yml) before merge.

Security / privacy

• Confirm no secrets/PII in this PR: Yes

---

🤖 Generated with Claude Code

[github-actions]

Overall Assessment

✅ Looks good — no unmitigated P0 or P1 findings.

Executive Summary

• The diff is release metadata only: version bumps and changelog promotion.
• No estimator, math, weighting, variance/SE, inference, or default behavior changed.
• Version surfaces are consistent across Python, packaging, Rust, citation metadata, and LLM guide.
• Methodology-related changelog notes point to already documented REGISTRY/TODO entries, so they are informational, not defects.
• No security, performance, or test blockers found.

Methodology

Finding: Documented methodology notes only

• Severity: P3 informational
• Impact: The changelog mentions LPDiD non-absorbing treatment, LPDiD survey support, CallawaySantAnna NaN materialization, and survey zero-weight PSU behavior, but this PR does not change implementation. The relevant deviations/limitations are documented in docs/methodology/REGISTRY.md:L1838-L1873, TODO.md:L95-L120, and TODO.md:L156.
• Concrete fix: None required.

Code Quality

No findings. Version bump is consistent in pyproject.toml:L7, diff_diff/__init__.py:L304, rust/Cargo.toml:L3, CITATION.cff:L10-L11, and diff_diff/guides/llms-full.txt:L5.

Performance

No findings. No executable performance-sensitive paths changed.

Maintainability

No findings. Changelog release header and comparison link were added at CHANGELOG.md:L10 and CHANGELOG.md:L1919.

Tech Debt

No findings. Deferred limitations referenced by release notes are already tracked in TODO.md.

Security

No findings. No secrets, credentials, network code, or dependency changes introduced.

Documentation / Tests

No blocking findings. This is a release-cut PR with documentation/version metadata only; no new tests are required for the changed surface. Static consistency checks found no stale 3.6.0 metadata outside historical changelog entries.