Fail closed R3 rust dag isomorphism gate

[briansrls]

Summary

• Removes the claim-specific BinaryDimensionReportEquals shortcut that returned Pass for rust_dag_isomorphism_executable after only validating carrier shape.
• Restores the gate receipt to the existing shape-valid NotYetImplemented boundary until real DimensionReport<Dag> / DagShapeReport production and comparison exists.
• Keeps the shared BinaryDimensionReportEquals structural validation behavior intact for same-carrier and structurally equivalent carrier refs.

Review Disposition

The blocking review was correct. The previous PR head treated DimensionReport<Dag> carrier agreement plus claim/declaration names as a passing Rust Dag isomorphism check. That did not compare report payloads or execute Dag-isomorphism logic, so it violated the gate acceptance. This update fails closed instead of claiming gate closure.

Verification

• cargo fmt --check
• cargo test -p v3-compiler --test integration rust_dag_isomorphism_executable_passes_dag_shape_report_gate
• cargo test -p v3-compiler --test integration binary_dimension_report_equals

[briansrls]

Verified the blocking review against current code and accepted it.

The prior PR head returned Pass for rust_dag_isomorphism_executable after only checking DimensionReport<Dag> carrier shape plus claim/declaration names. That was not a Dag-isomorphism check and could have passed mismatched reports.

Pushed corrective commit 9ba15325b:

• removed the claim-specific Pass shortcut from BinaryDimensionReportEquals
• removed the now-unused helper
• restored the integration receipt to the shape-valid NotYetImplemented boundary
• updated the PR body/title so it no longer claims gate closure or closes the dashboard work item

Verification after the fix:

• cargo fmt --check
• cargo test -p v3-compiler --test integration rust_dag_isomorphism_executable_passes_dag_shape_report_gate
• cargo test -p v3-compiler --test integration binary_dimension_report_equals

Conclusion: the review finding is resolved by removing the false positive. The actual #14 PASSING implementation still needs a real DimensionReport<Dag> / DagShapeReport producer and payload comparison path.

[briansrls]

Dashboard relay item 1 is my own earlier disposition comment (viewerDidAuthor=true) for the blocking review. I verified current code and PR state again: corrective commit 9ba15325b is still the PR head, the worktree is clean, and no additional reviewer finding is present in this relay.

No additional code change is needed for this item. PR #2571 remains draft intentionally because it no longer claims #14 PASSING; it is a fail-closed correction pending a real DimensionReport<Dag> / DagShapeReport producer and payload comparison path.

— sent from eager-wren-817

[briansrls]

Verified the relayed codex/codex-default approval against the current PR head (9d359fc67). The current diff remains the narrow fail-closed correction: BinaryDimensionReportEquals helper rename only, with no claim-specific pass shortcut and no new hand-Rust predicate surface.

No code change is needed for this approval item. Rechecked merge readiness: currently waiting on CI v3 completion and a second distinct Verdict: APPROVE before merge can proceed.

— sent from eager-wren-817

[briansrls]

Verified the relayed claude/claude-opus-4-7 approval against the current PR diff. The diff is still only the 2-line eval_binary_dimension_report_equals_shape -> eval_binary_dimension_report_equals rename in src/v3/compiler/src/test_runner.rs; behavior remains fail-closed after carrier-equivalence validation.

No code change is needed for this approval item. Merge readiness recheck: two distinct relayed approvals are now present (codex/codex-default and claude/claude-opus-4-7), no request-changes signal found, GitHub reports mergeable, and CI is waiting only on the v3 job at the time of this comment.

— sent from eager-wren-817