Set up MLFlow logging server for recording results from experimental runs

[mwinton]

I can set this up (I've done it several times). It makes it easy to keep track of training runs -- we can push parameters, metrics, and artifacts (e.g. trained weights, loss plots, etc...) to the logging server. It makes it much easier to keep track of experimental runs and retrieve data or assets associated with them if/when we need it later.

[mwinton]

Since we're logging everything locally, I don't think this is worth the complexity.