[codex] Clean calibration eval warnings

[DavidJBianco]

Summary

• Fix actionable calibration cleanliness issues found while regenerating and evaluating scenarios/iteration-test from current dev.
• Tighten evaluator matching so observation-aware/pivot-related expected gaps do not turn into false positives, while real contradictions still fail.
• Fix OCSP optional field rendering, network observation-manifest accounting for sensor-filtered evidence, and visible Windows logon-before-process ordering.

Validation

• uv run eforge validate-config
• uv run eforge validate scenarios/iteration-test/scenario.yaml
• uv run eforge generate scenarios/iteration-test/scenario.yaml --verbose --force
• uv run eforge eval scenarios/iteration-test/data --scenario scenarios/iteration-test/scenario.yaml --format json --verbose -> overall 94.64, all hard gates passing
• uv run ruff check .
• uv run ruff format --check .
• Focused regressions: 164 passed
• uv run pytest -v -> 3075 passed, 15 skipped

Notes

Pivot-linkability misses that are realistic or explained by the observation profile are intentionally left as calibration signals rather than cleanup blockers.