feat(ci): attach test and repro log artifacts to PR orchestrator runs

[djm81]

Description

Attach test and repro log artifacts to PR Orchestrator workflow runs so failed CI can be debugged by downloading full logs from the Actions Artifacts section without re-running locally.

• Tests job: Runs hatch run smart-test-full and uploads logs/tests/ as test-logs artifact; coverage XML still uploaded for quality gates.
• Contract-first-ci job: Captures specfact repro stdout/stderr to logs/repro/, uploads repro-logs and repro-reports (.specfact/reports/enforcement/) as artifacts.
• Docs: New "CI and GitHub Actions" section in Troubleshooting describing artifact names and how to use them.

Fixes #260

New Features (none)

Contract References: None (workflow and docs only). OpenSpec change: ci-01-pr-orchestrator-log-artifacts.

Type of Change

Please check all that apply:

• 🐛 Bug fix (non-breaking change which fixes an issue)
• ✨ New feature (non-breaking change which adds functionality)
• 💥 Breaking change (fix or feature that would cause existing functionality to not work as expected)
• 📚 Documentation update
• 🔒 Contract enforcement (adding/updating @icontract decorators)
• 🧪 Test enhancement (scenario tests, property-based tests)
• 🔧 Refactoring (code improvement without functionality change)

Contract-First Testing Evidence

Required for all changes affecting CLI commands or public APIs:

No application code or public API changes; workflow YAML and docs only.

Contract Validation

• Runtime contracts added/updated (@icontract decorators on public APIs)
• Type checking enforced (@beartype decorators applied)
• CrossHair exploration completed: hatch run contract-test-exploration
• Contract violations reviewed and addressed

Test Execution

• Contract validation: hatch run contract-test-contracts ✅
• Contract exploration: hatch run contract-test-exploration ✅
• Scenario tests: hatch run contract-test-scenarios ✅
• Full test suite: hatch run contract-test-full ✅

Test Quality

• CLI commands tested with typer test client — N/A (no CLI changes)
• Edge cases covered with Hypothesis property tests
• Error handling tested with invalid inputs
• Rich console output verified manually or with snapshots

How Has This Been Tested?

Contract-First Approach: Workflow and docs only; contract-test and smart-test run; workflow validated with hatch run lint-workflows and hatch run yaml-lint.

Manual Testing

• Workflow syntax validated with lint-workflows
• YAML lint passed
• Tested CLI commands manually — N/A
• Verified rich console output — N/A

Automated Testing

• Contract validation passes
• All existing tests still pass
• Workflow lint passes

Test Environment

• Python version: 3.12
• OS: Ubuntu (GitHub Actions runner)

Checklist

• My code follows the style guidelines (PEP 8, ruff format, isort)
• I have performed a self-review of my code
• I have added/updated contracts (@icontract, @beartype) — N/A
• I have added/updated docstrings (Google style) — N/A (no Python changes)
• I have made corresponding changes to documentation
• My changes generate no new warnings (basedpyright, ruff, pylint)
• All tests pass locally
• I have added tests that prove my fix/feature works — N/A (workflow/docs)
• Any dependent changes have been merged — PR to dev exists (feat(ci): attach test and repro log artifacts to PR orchestrator runs #261)

Quality Gates Status

• Type checking ✅ (hatch run type-check)
• Linting ✅ (hatch run lint)
• Contract validation ✅ (hatch run contract-test-contracts)
• Contract exploration ✅ (hatch run contract-test-exploration)
• Scenario tests ✅ (hatch run contract-test-scenarios)

Screenshots/Recordings (if applicable)

N/A — workflow and documentation changes only.