Dynamic Time Series Anomaly Detection Pipeline

Design a comprehensive Python script using pandas and numpy that automatically detects statistical anomalies in time series data with multiple configurable detection methods. The solution should support Z-score, Interquartile Range (IQR), and machine learning-based isolation forest techniques. Implement robust error handling, generate detailed anomaly reports with visualization, and create a modular architecture that can be easily integrated into existing data processing workflows. Include performance optimization techniques for handling large datasets and demonstrate how to configure detection sensitivity thresholds dynamically.