features to add

[EricSchles]

TODO: add a way to check for secondary metrics which make it easier to test for the realization of certain emergent properties of a machine learning model. This can be as naive as shap values, but should probably be other measures as well.

TODO: add a way to tell which train test splits produce different results. Give a way to interrogate the data further for the best and worst performing train test splits

TODO: add some of the testing that's talked about in https://www.youtube.com/watch?v=BDEBF62iZx0&ab_channel=PyData