Research consolidation: authenticate PR279's complete parent inventory (no new kappa)

[adamgall]

Research consolidation — no new kappa

Resolves the explicit batch970 request to reconcile PR279's complete 116-file parent inventory with its original reviewed source and incorporated #828 companion. Source head: 913ccc48a0bc1173709ea92280b96252256b7754; research base: 1b59b624d21924f8e3537cdc1ac52d3be6eda2fe; OS-random assignment seed 97.

New evidence, not a duplicate replay

#828 (mihir-s-05) already reconstructed the later package and ran its nine-stage finite verifier. This PR reuses that work explicitly; it does not claim a new mathematical replay. It adds:

• an offline proof authenticating the raw reviewed/current Git commits and 27 Git trees, binding every source byte and file mode to the fixed commit IDs;
• an exact check of all 104 retained + 12 companion = 116 parent paths, including every previously matched path/blob rather than only that list's size;
• a repository-wide, commit-to-commit Merkle-tree difference: exactly 10 modified + 2 added files, precisely the existing companion update, with no additional change hidden outside the assumed package;
• one positive case and 17 corruption/omission/substitution controls;
• an updated ledger priority linking the resolved source-coverage question while preserving the parent's disposition and all original frozen sources.

Both original commits, source archives, selected frontier and predecessor receipts remain unchanged. PR279's acceptance/closure remains a maintainer decision; this PR does not close it.

Reproduce and observed verification

python3 -B research/pr279-parent-closure/verify.py
python3 -I -O -B research/pr279-parent-closure/test_verify.py

Observed on Python 3.14.4: PASS_PR279_PARENT_COMMIT_CLOSURE; parent_paths=116, retained_paths=104, companion_paths=12, authenticated_commits=2, authenticated_trees=27; all 18 tests passed (Ran 18 tests in 17.764s, OK). No Git executable, network, .git directory, SymPy, or expensive mathematical replay is needed by these checks. A separate deterministic collector is included to recapture the frozen proof from original Git objects.

Exact report, source pins, limits, three compared candidates and final overlap refresh are under research/pr279-parent-closure/. The nine reused mathematical stages remain explicitly marked fresh_execution=false. No new selected-frontier cold replay, Lean build or hosted CI execution is claimed. Git SHA-1 source identity is a trust assumption, not an authorship/signature proof. All inherited compiler/geometry/restoration/routing/precision/analytic assumptions remain.

Credit and coordination

Adam Gall (@adamgall), with substantial OpenAI Codex assistance. Predecessors: mihir-s-05 (#828) for reconstruction/replay; Rohan Arun (#279), eumemic (#268), DreamingOfClouds (#276), Douglas Colkitt and the retained full source lineage. Original Apache-2.0 licenses and assistance disclosures remain intact.

Claim: #1166 (comment)

Distinct from PR116 evidence-only reconciliation (#1007), selected PR1042 capsule closure (#1142), and PR225 pricing consolidation (#1110).

[Mister-Meeseeks]

100-PR batch review at 24777831784d18896d74ea0c0ed12288d08c89b4.

Merged to main, preserving the original contributor commits and package attribution.

Reviewed scope: Commit/tree-authenticated 116-file PR279 parent inventory closure and exact repository delta; authenticates retained replay evidence without claiming fresh mathematics.

The pinned scoped command passed; the batch ledger contains its command, input hashes, environment and log. This incorporates the research/tooling contribution and does not change the selected κ or discharge the inherited multiplication-theorem assumptions.

Batch review, attribution and evidence. Thank you for the contribution.