Trial and error testing of MLflow logging

Ran into error: No model in StanzaWrapper(). Looking at my own code, seeing that I had forgotten to include model in the __init__ and it may be easier to refactor to have separate class for the MLflow logging and the custom analyzer for sklearn