Add Deep Learning (LSTM) + Quick Wins automation features

Major improvements:
- Deep Learning integration with PyTorch LSTM (Bidirectional, 128→64 units)
- Hybrid predictor: 40% RandomForest + 60% Deep Learning
- LaunchAgent for automatic weekly tip generation (Tue/Fri 21:00)
- Health-Check system with auto-recovery and Telegram alerts
- Model caching and intelligent retraining logic
- Updated CSV data and generated tips
- Performance reports for recent draws

Technical details:
- PyTorch used instead of TensorFlow (Python 3.14 compatibility)
- Apple Silicon MPS acceleration support
- Sequence learning with 20-draw history
- Early stopping and learning rate scheduling

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

data/AlleLottozahlen.csv

@@ -4967,3 +4967,9 @@ datum;Z1;Z2;Z3;Z4;Z5;Z6;SZ
 2025-12-06;15;26;27;33;35;37;2.0
 2025-12-10;3;11;12;13;17;24;8.0
 2025-12-13;6;11;21;32;37;45;3.0
+2025-12-17;6;13;25;26;32;37;2.0
+2025-12-20;12;23;32;33;40;42;5.0
+2025-12-24;7;8;25;32;37;39;8.0
+2025-12-27;7;10;27;28;38;46;6.0
+2025-12-31;3;5;7;28;36;41;5.0
+2026-01-03;3;8;11;12;16;46;8.0