Model support

[mmschlk]

Improvements and Ease of Use

• adds a class_label parameter to TabularExplainer and Explainer to specify the class label to be explained for classification models #271
• adds support for PyTorch models to Explainer #272

Bug Fixes

• fixes a bug that RandomForestClassifier models were not working with the TreeExplainer #273

The What I changed:

• main changes are done to explainer.utils where I add a wrapper callable which extracts class_index values from the prediciton functions.
• fix for RandomForest is simple, moving where the scaling of each tree is applied down after the values are potentially transformed into probabilities
• added a lot of additional tests checking for the efficency property, checking that shapiq can handle different model classes (more soon) and adding comparison of shapiq to shap