feat(roster): add capability suggestions and receipt stats

[solomonneas]

Summary

• Add brigade roster suggest --preset NAME --target PATH on top of the resolver merged in feat(roster): add shareable presets and fallback resolution #454.
• Add local per-seat duration medians and failure rates from worker receipts.
• Add capability warnings and stats lines to roster doctor without changing its warning-only exit behavior.
• Emit adoptable roster TOML only when the resolved orchestrator remains available.

The implementation keeps the existing four packaged presets unchanged and does not add a second resolver module. Local receipt fields replace author stats in emitted TOML, so author values are not relabeled as local evidence.

This supersedes #462, which duplicated the resolver and preset work already merged in #454.

Verification

• pytest -q tests/test_roster.py tests/test_roster_cmd.py: 159 passed in 1.09s
• ./scripts/verify: 4071 passed, 3 skipped in 381.90s, 82.53% coverage

Closes #444.

[coderabbitai]

Important

Review skipped

Auto reviews are limited based on label configuration.

🏷️ Required labels (at least one) (1)

• coderabbit-review

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Repository: escoffier-labs/coderabbit/.coderabbit.yaml

Review profile: CHILL

Plan: Pro Plus

Run ID: 3b79a82e-d471-4199-82fd-3bea116988d1

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch codex/issue-444-parts-3-4-final

---

_{Comment @coderabbitai help to get the list of available commands.}