Feature: Machine learning interface for direction reconstruction
- Adds an interface to run machine learning models (e.g., scikit-learn).
- Provides an example using XGBoost regression trees to estimate directions in camera coordinates.
- Introduces a Python environment and a scripts directory to support training and inference.
The new XGBoost-based training and reconstruction methods estimate the gamma‑ray arrival direction in camera coordinates. They support DispBDT and intersection methods, combining image parameters to compute Xoff/Yoff.
Compared to previous approaches that averaged over all DispBDT directions, this implementation better handles edge cases (e.g., truncated images).