Fix multiple critical bugs in onboarding training step

This commit addresses all identified bugs and issues in the training code path:

## Critical Fixes:
- Add get_start_time() method to TrainingLogRepository and fix non-existent method call
- Remove duplicate training.started event from API endpoint (trainer publishes the accurate one)
- Add missing progress events for 80-100% range (85%, 92%, 94%) to eliminate progress "dead zone"

## High Priority Fixes:
- Fix division by zero risk in time estimation with double-check and max() safety
- Remove unreachable exception handler in training_operations.py
- Simplify WebSocket token refresh logic to only reconnect on actual user session changes

## Medium Priority Fixes:
- Fix auto-start training effect with useRef to prevent duplicate starts
- Add HTTP polling debounce delay (5s) to prevent race conditions with WebSocket
- Extract all magic numbers to centralized constants files:
  - Backend: services/training/app/core/training_constants.py
  - Frontend: frontend/src/constants/training.ts
- Standardize error logging with exc_info=True on critical errors

## Code Quality Improvements:
- All progress percentages now use named constants
- All timeouts and intervals now use named constants
- Improved code maintainability and readability
- Better separation of concerns

## Files Changed:
- Backend: training_service.py, trainer.py, training_events.py, progress_tracker.py
- Backend: training_operations.py, training_log_repository.py, training_constants.py (new)
- Frontend: training.ts (hooks), MLTrainingStep.tsx, training.ts (constants, new)

All training progress events now properly flow from 0% to 100% with no gaps.

frontend/src/components/domain/onboarding/steps/MLTrainingStep.tsx

@@ -5,6 +5,11 @@ import { Button } from '../../../ui/Button';
 import { useCurrentTenant } from '../../../../stores/tenant.store';
 import { useCreateTrainingJob, useTrainingWebSocket, useTrainingJobStatus } from '../../../../api/hooks/training';
 import { Info } from 'lucide-react';
+import {
+  TRAINING_SKIP_OPTION_DELAY_MS,
+  TRAINING_COMPLETION_DELAY_MS,
+  SKIP_TIMER_CHECK_INTERVAL_MS
+} from '../../../../constants/training';
 
 interface MLTrainingStepProps {
   onNext: () => void;
@@ -38,16 +43,16 @@ export const MLTrainingStep: React.FC<MLTrainingStepProps> = ({
   const currentTenant = useCurrentTenant();
   const createTrainingJob = useCreateTrainingJob();
 
-  // Check if training has been running for more than 2 minutes
+  // Check if training has been running for more than the skip delay threshold
   useEffect(() => {
     if (trainingStartTime && isTraining && !showSkipOption) {
       const checkTimer = setInterval(() => {
         const elapsedTime = (Date.now() - trainingStartTime) / 1000; // in seconds
-        if (elapsedTime > 120) { // 2 minutes
+        if (elapsedTime > TRAINING_SKIP_OPTION_DELAY_MS / 1000) {
           setShowSkipOption(true);
           clearInterval(checkTimer);
         }
-      }, 5000); // Check every 5 seconds
+      }, SKIP_TIMER_CHECK_INTERVAL_MS);
 
       return () => clearInterval(checkTimer);
     }
@@ -72,14 +77,14 @@ export const MLTrainingStep: React.FC<MLTrainingStepProps> = ({
       message: 'Entrenamiento completado exitosamente'
     });
     setIsTraining(false);
-    
+
     setTimeout(() => {
       onComplete({
         jobId: jobId,
         success: true,
         message: 'Modelo entrenado correctamente'
       });
-    }, 2000);
+    }, TRAINING_COMPLETION_DELAY_MS);
   }, [onComplete, jobId]);
 
   const handleError = useCallback((data: any) => {
@@ -147,7 +152,7 @@ export const MLTrainingStep: React.FC<MLTrainingStepProps> = ({
           message: 'Modelo entrenado correctamente',
           detectedViaPolling: true
         });
-      }, 2000);
+      }, TRAINING_COMPLETION_DELAY_MS);
     } else if (jobStatus.status === 'failed') {
       console.log(`❌ Training failure detected (source: ${isConnected ? 'WebSocket' : 'HTTP polling'})`);
       setError('Error detectado durante el entrenamiento (verificación de estado)');
@@ -169,13 +174,15 @@ export const MLTrainingStep: React.FC<MLTrainingStepProps> = ({
     }
   }, [jobStatus, jobId, trainingProgress?.stage, onComplete, isConnected]);
 
-  // Auto-trigger training when component mounts
+  // Auto-trigger training when component mounts (run once)
+  const hasAutoStarted = React.useRef(false);
   useEffect(() => {
-    if (currentTenant?.id && !isTraining && !trainingProgress && !error) {
+    if (currentTenant?.id && !hasAutoStarted.current && !isTraining && !trainingProgress && !error) {
       console.log('🚀 Auto-starting ML training for tenant:', currentTenant.id);
+      hasAutoStarted.current = true;
       handleStartTraining();
     }
-  }, [currentTenant?.id]); // Only run when tenant is available
+  }, [currentTenant?.id, isTraining, trainingProgress, error]); // Include all checked dependencies
 
   const handleStartTraining = async () => {
     if (!currentTenant?.id) {