Per-visual cluster performance/confusion matrix

[WarmCyan]

It would be useful to be able to view performance metrics such as class breakdown and score of the model on each visual cluster, similar to the visual cluster frequent/important words views. This would likely best be represented with individual confusion matrices.

(It can be difficult to determine accuracy within a cluster just from the "focus misclassifications" when a cluster is particularly dense)