Centralize sensitive log redaction across services

[quarj0]

Closes #150

Summary

• adds one shared backend redaction processor for Django, Celery, storage/provider logging paths, and managed AI
• installs the boundary before Django/Celery/AI application logging can emit records
• sanitizes nested structured context, positional args, exception text/tracebacks, credentials, common PII shapes, evidence locations, biometric fields, and binary values
• adds frontend safeLog and rejects direct production console.* calls in lint
• adds adversarial cross-service tests and operational logging guidance
• copies the shared backend logging package into both production images

Verification

• DJANGO_SETTINGS_MODULE=config.settings.testing uv run --project backend python backend/django/manage.py test common
• cd backend/ai-service && uv run --project .. pytest tests/test_safe_logging.py
• cd frontend && pnpm --filter @identitycore/api-client lint
• full CI and Security workflows

Security notes

• safe operational identifiers/status/reason codes remain available
• credentials, subject PII, document/OCR content, evidence storage keys, selfies/embeddings/liveness media, and byte payloads are redacted
• exception strings are sanitized at the Python LogRecord creation boundary so callers cannot bypass redaction by using logger.exception(...) or passing an exception as a formatting argument

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, you can upgrade your account or add credits to your account and enable them for code reviews in your settings.