Retire conflict Bats and Python cases covered by Go tests #906

Problem

The workflow suite at tests/skills/resolve-rebase-conflicts/scripts/resolve-workflow.bats:7 and classifier suite at tests/skills/resolve-rebase-conflicts/scripts/test_conflict_shape.py:61 protect different contracts. Workflow contexts, refusal statuses and interpreter metadata cannot all be replaced by a union-output test. Retire them separately from production ports with explicit coverage evidence.

Part of #588 and #326.

Acceptance criteria

1. AC1: Given all workflow and Python cases including parameterized rows, when the ledger is read, then every case is mapped to a covering Go test or a reason to retain it.
2. AC2: Given a deleted case, when its mapping is inspected, then its still-shipped assertions have equivalent Go coverage.
3. AC3: Given interpreter or fallback prerequisite cases still needed during N-2, when the suites are reduced, then those cases remain unless equivalent Go-driven fallback coverage exists.
4. AC4: Given the retirement diff, when inspected, then it removes only mapped duplicate tests and their unused scaffolding.
5. AC5: Given the ledger, when covering tests are inspected, then each identifies its coverage kind and actual executing platforms.
6. AC6: Given a promised platform, when the covering tests run in CI, then they execute without skips and pass.
7. AC7: Non-functional: retirement changes no resolver implementation, refusal rule or frozen fallback.

Scope and non-goals

Scope: conflict workflow Bats and helper pytest coverage, plus a retirement ledger.

Non-goal: removing production Python, removing shell, treating native execution without Python as coverage for a legacy missing-interpreter diagnostic, or retiring PreToolUse cases. Legacy-algorithm cases still lacking Go coverage block child 15. Finish their separate test retirement here once coverage exists; no Bats or pytest retirement is bundled into production removal. Any case retained beyond removal must cover a surviving shell entry.

Approach

Map the seven Bats cases and every Python test/parameter row, including mode misuse, exit 3 versus defects, manual declarations, binary/empty input, token proposals and PEP 723 metadata. Use the Go ports from children 05 to 09 and activation coverage from child 10. Interpreter requirements at tests/skills/resolve-rebase-conflicts/scripts/test_conflict_shape.py:206 remain true of the frozen helper until its later removal. Keep their tests or cover those exact artifacts from Go. Do not use an 'obsolete' label to delete a still-applicable assertion.

Dependencies

Blocked by: #899, #900, #901, #902, #903, #904.

No dependency on N-2 for cases already covered by equivalent Go tests.

Size

S. Forecast 330 code lines: up to 270 measured test code lines deleted and 60 ledger-check/coverage adjustments. Prose allowance: 280 lines. Substantive new behavior must land in an earlier Go child.

scc could not run in the isolated checkout. Measured baselines count non-blank lines other than full-line # comments, including Python docstrings and embedded code conservatively. Forecasts include implementation, tests, fixtures and registration; they are estimates, not measured future Go. Generated output is excluded only under packages/agents/claude/.apm/skills/, packages/agents/codex/.apm/skills/, .claude/skills/, .agents/skills/ and tokens.json. The source under packages/agents/source/ and copies anywhere else count. The final diff must be at most 800 code lines and 2,000 prose lines, with no exception; split before implementation if either forecast grows beyond its limit.

Verify plan

• AC1, AC2: enumerate Bats cases and Python parameterized rows and inspect each assertion-level Go mapping.
• AC3: audit missing-uv, missing-python, relative-entry and metadata cases against the still-shipped frozen helper.
• AC4, AC7: inspect the diff for test-only changes and confirm fallback hashes remain unchanged.
• AC5, AC6: inspect coverage kind, platform promises and CI logs; run any retained Bats/pytest cases on the platforms named in the ledger.
• Definition of done: D1 each AC is verified at the merged head, with the commands and results cited in the pull request's Verification. D2 an independent review approves the exact merged head, and every finding has a recorded disposition. D3 required checks pass with no gate loosened. D4 the leak scans the repository specifies are clean, and every post made for this work passes the per-post leak review before it is posted. D5 every commit is signed. D6 the documentation matches the change or the Verification says why there is no impact. D7 every commit has an Assisted-by trailer and the pull request has the provenance:agent-assisted label. D8 the pull request stays within the size limits. D9 follow-ups are filed as issues and listed under Related. D10 the merge closes the issue through a closing reference, the board shows Done, and the final Status: comment names the merge commit.

Risks

Deleting interpreter or exit-contract cases too early can break the only compatibility path older consumers have. Undo: revert this test-only commit.

Operator authority

None.