- Add user_id column to TrainingRun model in db.py
- Store user_id on TrainingRun insert in /training/start
- Filter GET /training/runs by user_id (returns empty list if no user context)
- Enforce user ownership on GET /training/runs/{run_id} (404 on mismatch)
- Enforce user ownership on DELETE /training/runs/{run_id} (404 on mismatch)
- Add migration 002 to add user_id column and index to training_runs table
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Remove hardcoded SQL comments containing 'ml_user' and 'ml_password'
- Remove fallback default credentials in DATABASE_URL construction
- Add fail-fast validation: raise RuntimeError if DATABASE_URL env var is missing or empty
- Mark task 1.4 as complete in code-review-fix/tasks.md
- Fix CCI indicator to use HLC prices instead of close only
- Parse datetime column when loading enriched CSV
- Strip timezone from annotation timestamps
- Fix TA-Lib pattern names (CDL3WHITESOLDIERS, CDL3BLACKCROWS)
- Exclude programmatic label columns from training features
- Fix classification report to handle missing classes
- Update MLflow tracking to use localhost:5000
- Grant PostgreSQL permissions to ml_user
Pipeline now runs successfully end-to-end:
- Feature engineering: 2543 rows, 31 columns
- Annotation ingestion: 286 samples
- Training: 89.47% test accuracy with Random Forest
