Investigate potential for using shared mutations for clustering

[matsen]

This paper finds that B cell phylogenies look like this:

This could be very helpful for BCR clustering, because we could use shared mutations to distinguish between lineages.

In a collection of the biggish clusters on real data, do we see this too?

This could be evaluated by either building trees or by looking at shared mutations in germline encoded regions.

[psathyrella]

Very similar to #29

[matsen]

Make this a second phase?