[Feature] Model-Based Feature Selection

[ErikBavenstrand]

Is your feature request related to a problem? Please describe.
Currently, the mleko library does not provide support for model-based feature selection. This is a problem as it leaves users with limited options for feature selection.

Describe the solution you'd like
I would like to see the integration of model-based feature selection methods in mleko. Specifically, the following methods would be beneficial:

1. Lasso (L1) regularization: This method can help in feature selection by shrinking the coefficients of less important features to zero.

2. Tree-based feature selection: Decision trees or tree-based ensemble methods (like Random Forests and Gradient Boosting) can be used to rank features based on their importance.

3. Recursive Feature Elimination (RFE): This is a greedy optimization algorithm which aims to find the best performing feature subset. It repeatedly creates models and keeps aside the best or the worst performing feature at each iteration.

Describe alternatives you've considered
An alternative could be to use other libraries or tools that provide model-based feature selection. However, having this functionality integrated within mleko would make the workflow more efficient.

Additional context
Model-based feature selection methods can provide more accurate and reliable feature selection, which can lead to better model performance. Therefore, adding these methods would greatly enhance the functionality of the mleko library.