Reorg the docs for the Interpretable Causal ML, Validation, Visualization

[paullo0106]

Describe the solution you'd like
As Jeong mentioned in #413, we can improve the documentation by consolidating Interpretable CausalML, Validation, Visualization

We can remove the Visualization section and split the existing content to Interpretable (feature importance, SHAP, tree plot) and Validation sections (AUUC and etc).