Centralize operational divergence classification

[Taz33m]

Summary

• define operational divergence once in the classification layer
• reuse it for in-memory reports and serialized artifact exit codes
• remove the classification module's runtime dependency on compare
• preserve every existing category/snapshot/close classification

Verification

• full pytest suite
• black, flake8, mypy, bandit, compileall
• exhaustive 297-case baseline comparison is byte-for-byte identical

---

Summary by cubic

Centralized operational divergence classification in classification.is_operational_divergence, and reused it for ConformanceReport.operational_failure and artifact exit codes. No behavior change; this removes classification’s runtime dependency on compare and keeps exit codes consistent.

• Refactors
  • Added is_operational_divergence(category, snapshot_failed, close_failed) in classification.
  • Used it in compare.ConformanceReport.operational_failure and exit_codes._divergence_exit_code.
  • Switched classification → compare import to type-only to avoid runtime coupling.
  • Kept all classifications unchanged; baseline remains identical and tests assert operational_failure.

^{Written for commit c874f4e. Summary will update on new commits.}

[cubic-dev-ai]

No issues found across 4 files

_{Re-trigger cubic}