Tighten offline tests and adoption diagnostics

[jeffhuber]

Summary

• make the standalone-wrapper reinstall fixture prove offline behavior with a stdlib-only local build backend
• identify packaged versus explicit configuration in setup-drift text and JSON
• surface the adoption posture hint once, before provider-specific warning detail
• document the resulting install and upgrade diagnostics

Verification

• focused tests: 5 passed
• full unit suite: 855 passed
• Ruff: passed
• compileall: passed
• privacy scan: passed
• easy-mode smoke: passed

Builder experiment

Muse CLI 1.0.2 completed the first implementation in 390.7 seconds. One orchestration correction was required for Muse CLI worktree argument syntax. Orchestrator review found one duplicate posture-hint rendering; Muse completed fix round 1 in 56.0 seconds and added regression coverage. Peer audit findings remain hidden until this PR opened.

Privacy: authoring artifacts contain metadata only; command output, source, diffs, transcripts, auth output, local paths, and secrets were not captured for upload.

Closes #656

CODE_MOWER_BUILDER:muse
CODE_MOWER_LANE:muse
CODE_MOWER_TASK_CLASS:package-runtime
CODE_MOWER_EXPERIMENT:v104-underrepresented-builder-calibration
CODE_MOWER_FINDING_DISPOSITION:accepted-fixed duplicate-posture-hint
CODE_MOWER_FIX_ROUNDS:1
CODE_MOWER_USER_INTERVENTIONS:1
CODE_MOWER_TESTS:pass
CODE_MOWER_PRIVACY:metadata-only
CODE_MOWER_END

[gitar-bot]

Code Review ✅ Approved

Tightens offline wrapper fixture to prove offline behavior with stdlib-only local build backend, identifies packaged versus explicit configuration in setup-drift text and JSON, and surfaces adoption posture hint once before provider-specific warnings. All tests pass with no issues found.

Options

Auto-apply is off → Gitar will not commit updates to this branch.
Display: compact → Showing less information.

Comment with these commands to change the behavior for this request:

Auto-apply	Compact
gitar auto-apply:on	gitar display:verbose

_{Was this helpful? React with 👍 / 👎 | Gitar}

[jeffhuber]

Claude audit (merge-authority lane)

Head SHA: 16bf3f79ccd20f4ea53a3efca548f2ab29c10b0b
Findings: P0=0, P1=0, P2=0, P3=0 (blocker policy: any P0/P1/P2 -> BLOCKED)

Claude Audit: PASS

Summary:

Additive, well-tested changes: doctor text rendering surfaces an adoption posture hint once (verified no duplication via group filtering and matching test), migration setup-drift report/text gains a non-breaking "config_source" field, docs updated to match, and test_release_hygiene.py swaps a setuptools fixture for a minimal offline PEP 517/660 backend plus PIP_NO_INDEX to keep the venv-reinstall test network-free. No data loss, security, or contract-breaking issues found.

Findings: none.

[jeffhuber]

Codex audit (merge-authority lane)

Head SHA: 16bf3f79ccd20f4ea53a3efca548f2ab29c10b0b
Findings: P0=0, P1=0, P2=0, P3=0 (blocker policy: any P0/P1/P2 → BLOCKED)

Codex Audit: PASS

Summary:

No actionable correctness regressions were identified.

Findings: none.