Map Auto Review proof metrics onto Codex review primitives

[shiny-code-bot]

Finish Line

Auto Review proof metrics and background review evidence are mapped to Codex review/guardian primitives with fixtures before review implementation changes resume.

Current Status

State: Created from #397 parity ledger.

Deleted evidence from the pre-#390 tree includes code-rs/core/src/review_coord.rs, code-rs/core/src/review_store.rs, code-rs/core/docs/auto-review.md, code-rs/core/tests/review_coord_integration.rs, TUI review settings, and Auto Drive review snapshots.

Current main has Codex review surfaces such as review/start, reviewer mode, guardian review events, and review formatting, but old Every Code proof ledger semantics are not yet classified.

Acceptance Criteria

• Diff old review_store.rs/review_coord.rs behavior against current Codex review/guardian surfaces.
• Classify each proof metric and background review guarantee as port, rewrite, covered, defer, or retire.
• Add parity fixtures for must-preserve proof artifacts before implementation ports.
• Do not rewrite current review coordination before the proof-metrics gap list is documented.
• End code-bearing slices with ./build-fast.sh passing cleanly.

Relationships

• Child of Stabilize Every Code parity on Codex CLI fork baseline #397.
• Related to Inventory and port Every Code features onto Codex base #386 feature inventory and prior proof-metrics work Add Auto Review proof metrics and dogfood diagnostics #330/Update Auto Review lifecycle docs for durable runs #331.

[cbusillo]

Closed by PR #414.

Validation completed before merge:

• python3 -m json.tool tools/code-exec-harness/scenarios/auto-review-config-routing.json >/dev/null
• python3 -m py_compile tools/code-exec-harness/harness.py
• git diff --check
• ./build-fast.sh
• python3 tools/code-exec-harness/harness.py tools/code-exec-harness/scenarios/auto-review-config-routing.json --code-bin /Users/cbusillo/.code/worktrees/code/auto-review-proof-metrics/code-rs/target/dev-fast/code
• CODE_EXEC_HARNESS_BIN=/Users/cbusillo/.code/worktrees/code/auto-review-proof-metrics/code-rs/target/dev-fast/code tools/code-exec-harness/run-deterministic.sh
• PR Add Auto Review proof metrics fixture #414 blob-size policy passed.

This slice deliberately validated before porting implementation behavior:

• Added docs/auto-review-proof-metrics-parity.md to classify old Auto Review proof metrics as covered/rewrite/defer/retire before deeper ports.
• Added a deterministic no-live-token harness scenario for the current covered primitive: approvals_reviewer = "auto_review" is accepted and emitted through the current wire-compatible guardian_subagent spelling on code exec --json thread.started.
• Exposed approvals_reviewer on the JSON thread-start event so future proof fixtures can assert reviewer routing directly.

Current broad Auto Review was not counted as validation for this slice. Ledger run 398c40ae-0323-4e2b-9e01-46fbb80e8d01 is current/terminal but failed before findings because the diff was 1,479,472 chars and the configured background review limit is 120,000 chars. That is now documented as a proof-source gap rather than treated as a merge gate.