Is your feature request related to a problem? Please describe.

ComparisonReport, which will be in v0.7, does not enable plots to easily benchmark several estimators

Describe the solution you'd like

As in Classification of text documents using sparse features, we could have:

As in Clustering text documents using k-means, we could have:

As in Sample pipeline for text feature extraction and evaluation, we could have:

Describe alternatives you've considered, if relevant

No response

Additional context

Note that this requires fit_time and the score_time