Stratified Sampling in Clustering Samplers

[JacksonBurns]

We already have most of the machinery in place to implement stratified sampling, the implementation might prove tricky (in terms of actually filling all the requested sets) but doable.

I welcome any literature references, comments on the utility, thoughts about the implementation, etc. Particular examples (especially Jupyter notebooks or vanilla scripts) containing code that demonstrate the utility of stratified sampling are especially welcomed.