fix(ml): complete ML pipeline fixes and setup

- 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

services/ml/app/db.py

@@ -15,6 +15,11 @@ from sqlalchemy.orm import sessionmaker, Session
 from sqlalchemy.sql import func
 
 
+# CREATE DATABASE ml_service;
+# CREATE USER ml_user WITH ENCRYPTED PASSWORD 'ml_password';
+# GRANT ALL PRIVILEGES ON DATABASE ml_service TO ml_user;
+
+
 # Database connection configuration from environment
 DATABASE_URL = os.getenv(
     "DATABASE_URL",
@@ -43,9 +48,9 @@ Base = declarative_base()
 # Training runs model
 class TrainingRun(Base):
     """Model for tracking ML training runs."""
-    
+
     __tablename__ = "training_runs"
-    
+
     id = Column(Integer, primary_key=True, index=True)
     run_id = Column(String(255), unique=True, nullable=False, index=True)
     model_type = Column(String(100), nullable=False)
@@ -56,7 +61,7 @@ class TrainingRun(Base):
     status = Column(String(50), nullable=False, default="running", index=True)
     created_at = Column(DateTime(timezone=True), server_default=func.now(), index=True)
     completed_at = Column(DateTime(timezone=True))
-    
+
     def __repr__(self):
         return f"<TrainingRun(run_id='{self.run_id}', status='{self.status}')>"
 
@@ -73,14 +78,14 @@ def init_db():
 def get_db() -> Generator[Session, None, None]:
     """
     Context manager for database sessions.
-    
+
     Usage:
         with get_db() as db:
             # Use db session here
             training_run = TrainingRun(run_id="123", ...)
             db.add(training_run)
             db.commit()
-    
+
     Yields:
         Database session
     """
@@ -94,12 +99,12 @@ def get_db() -> Generator[Session, None, None]:
 def get_db_session() -> Session:
     """
     Get a database session (for dependency injection).
-    
+
     Usage with FastAPI:
         @app.get("/")
         def endpoint(db: Session = Depends(get_db_session)):
             # Use db here
-    
+
     Returns:
         Database session (caller must close it)
     """