Align contributor test guidance with stdlib unittest

[CoreyRDean]

Non-technical summary

This tightens the contributor path around test execution so the instructions match what clauck actually supports in a clean checkout. Right now two test modules tell contributors to use pytest, but the repo and CI both rely on the Python standard library unittest runner instead.

This matters now because a fresh environment can successfully validate the repo without installing extra tooling, but the stale guidance sends contributors down a failing path first.

Technical summary

• Updated the module-level Run: guidance in tests/test_clauck.py to use python3 -m unittest tests.test_clauck
• Updated the module-level Run: guidance in tests/test_clauck_cron.py to use python3 -m unittest tests.test_clauck_cron
• Added the full unit-suite command to CONTRIBUTING.md so the documented validation path matches CI and the repository's minimal-dependency contract
• Preserved the existing scheduler-specific and shell syntax verification guidance
• No breaking changes

Validation

• Ran /usr/bin/python3 -m unittest discover tests -v

Additional notes

Trade-off: this is intentionally scoped to the misleading invocation surfaces rather than broad contributor-doc cleanup.

Deferred follow-up: none required inside this slice; broader documentation rework would be a separate increment if needed.

Remaining gap to fuller vision: the repo still relies on scattered inline Run: hints across tests, so future drift is still possible unless that guidance is centralized later.