Add Gradient Boosting model for stock prediction

[Aash1722]

Hi 👋,
It’s a great collection of finance and ML examples.

I noticed that the machine_learning section already includes ARIMA, LSTM, neural networks, clustering, and regression-based approaches. One method that seems missing and could complement the existing models is Gradient Boosting (e.g., sklearn’s GradientBoostingRegressor), which is widely used for tabular financial time-series data.

In addition, I was wondering if the project would be open to a simple hybrid approach, such as:

• Using LSTM to capture temporal patterns
• Feeding LSTM predictions or extracted features into a boosting model (GB / XGBoost-style)

I’d like to contribute a new script that:

• Uses historical stock data (same data source/style as existing scripts)
• Engineers financial features (returns, volatility, momentum)
• Predicts next-day returns
• Includes RMSE evaluation and visualization
• Follows the same interactive and educational style as lstm_prediction.py

If this sounds useful, I’m happy to open a PR with a clean, well-documented implementation under machine_learning/.

Please let me know your thoughts — I’d love to contribute!

Thanks again.