Add historical training data logging for hybrid models

Integrate saving of training history to the database within `train_hybrid.py`, ensuring that model versioning is correctly applied for hybrid detector runs.

Replit-Commit-Author: Agent
Replit-Commit-Session-Id: 7a657272-55ba-4a79-9a2e-f1ed9bc7a528
Replit-Commit-Checkpoint-Type: full_checkpoint
Replit-Commit-Event-Id: 9f8d0aa1-70ec-4271-b143-5f66d1d3756b
Replit-Commit-Screenshot-Url: https://storage.googleapis.com/screenshot-production-us-central1/449cf7c4-c97a-45ae-8234-e5c5b8d6a84f/7a657272-55ba-4a79-9a2e-f1ed9bc7a528/RJGlbTt

python_ml/train_hybrid.py

@@ -60,6 +60,44 @@ def train_on_real_traffic(db_config: dict, days: int = 7) -> pd.DataFrame:
     return df
 
 
+def save_training_history(db_config: dict, result: dict):
+    """
+    Save training results to database training_history table
+    """
+    import psycopg2
+    
+    MODEL_VERSION = "2.0.0"  # Hybrid ML Detector version
+    
+    print(f"\n[TRAIN] Saving training history to database...")
+    
+    try:
+        conn = psycopg2.connect(**db_config)
+        cursor = conn.cursor()
+        
+        cursor.execute("""
+            INSERT INTO training_history 
+            (model_version, records_processed, features_count, training_duration, status, notes)
+            VALUES (%s, %s, %s, %s, %s, %s)
+        """, (
+            MODEL_VERSION,
+            result['records_processed'],
+            result['features_selected'],  # Use selected features count
+            0,  # duration not implemented yet
+            'success',
+            f"Anomalie: {result['anomalies_detected']}/{result['unique_ips']} - {result['model_type']}"
+        ))
+        
+        conn.commit()
+        cursor.close()
+        conn.close()
+        
+        print(f"[TRAIN] ✅ Training history saved (version {MODEL_VERSION})")
+        
+    except Exception as e:
+        print(f"[TRAIN] ⚠ Failed to save training history: {e}")
+        # Don't fail the whole training if just logging fails
+
+
 def train_unsupervised(args):
     """
     Train unsupervised model (no labels needed)
@@ -71,6 +109,9 @@ def train_unsupervised(args):
     
     detector = MLHybridDetector(model_dir=args.model_dir)
     
+    # Database config for later use
+    db_config = None
+    
     # Load data
     if args.source == 'synthetic':
         print("\n[TRAIN] Using synthetic dataset...")
@@ -109,6 +150,10 @@ def train_unsupervised(args):
     print(f"  Model type:          {result['model_type']}")
     print("="*70)
     
+    # Save training history to database (if using database source)
+    if db_config and args.source == 'database':
+        save_training_history(db_config, result)
+    
     print(f"\n✅ Training completed! Models saved to: {args.model_dir}")
     print(f"\nNext steps:")
     print(f"  1. Test detection: python python_ml/test_detection.py")