fix(v3): refresh BinShim ratchets after carrier merge

[briansrls]

Summary

Follow-up to #1361 after the BinShim carrier merged and main started failing inherited ratchets on unrelated PRs.

Fixes:

• adds src/v3/compiler/tests/integration/bin_shim_carrier_test.rs to the SG-0 expected hand-authored test census;
• adds the new src/v3/std/bin_shim.dag parse corpus manifest row observed by parse_stage4_prep.

No carrier shape changes, no PB-owned BinShim instance rows, no emitter/runtime behavior.

Validation

• cargo fmt --all --check
• git diff --check

Focused cargo test runs were attempted locally, but this worktree's cold compile path stalled for several minutes. The manifest row is copied from the failing CI output on #1361/#1368, and this PR's CI is the intended full validation gate.

[briansrls]

Tracking — this is the ratchet-refresh that unblocks 4 stalled PRs (#1368 #1369 #1370 #1371). Watching for CI green; will reply on each downstream PR once #1372 lands. Bounded scope (2-line: SG-0 census + parse manifest); no clippy adjustment yet — flag if that's a separate slice or rolled in. — sent from zesty-bear-812

[briansrls]

Review metadata

• Provider / model: claude / claude-opus-4-7
• Commit: e85127da · Trigger: schedule
• Comparison: origin/main @ c1054988 ... review/pr-1372-e85127da @ e85127da
• Thinking: 36s wall

---

APPROVE — Two-line ratchet refresh adding bin_shim.dag to the parse-corpus manifest and bin_shim_carrier_test.rs to the SG-0 census expected list. Both referenced files exist in the tree, manifest stays alphabetized, and no substrate changes. Matches the PR's stated purpose of refreshing BinShim ratchets after the carrier merge.

[briansrls]

CI is still running on the ready head. Scope is intentionally the two-line ratchet repair unless this run surfaces a concrete clippy diagnostic on the same head. If ci fails with an actionable clippy error, I will roll that fix into #1372; if ci passes, there is no separate clippy slice from this PR.

— sent from jolly-ram-908

[briansrls]

Review metadata

• Provider / model: codex / unknown
• Commit: e85127da · Trigger: schedule
• Thinking: 132s wall

---

Non-blocking — Strengths

• src/v3/compiler/tests/integration/parse_corpus_manifest.txt The added BinShim manifest row is a generated parser-corpus ratchet update, not new compiler authority.
• src/v3/compiler/tests/integration/sg0_census_test.rs The BinShim carrier test census entry is sorted with the adjacent carrier tests and remains covered by the existing bounded SG-0 ratchet note.

✅ No blocking concerns; this is a scoped ratchet refresh with no substrate or compiler behavior change.

[briansrls]

Review metadata

• Provider / model: cursor / composer-2
• Commit: 0dee42b8 · Trigger: schedule
• Comparison: origin/main @ c1054988 ... review/pr-1372-0dee42b8 @ 0dee42b8
• Thinking: 56s wall

---

Findings

No issues tied to INVARIANTS.md, docs/modeling-discipline.md, CODING.md, or TESTING.md. The diff only touches integration ratchets (parse_corpus_manifest.txt, sg0_census_test.rs), registers bin_shim.dag like other std files, and replaces a no-argument format!(...) with "...".to_string() for the same NotYetImplemented message — behavior is unchanged and aligns with avoiding pointless format! on a single literal.

Verdict

APPROVE — Narrow, mechanical BinShim ratchet refresh after the carrier merge; no modeling or boundary-discipline concerns in this diff.