Cohort Pathways - Add statistics for time from one cohort to next

[davekern]

Expected behavior

Add statistics on the timing of the event cohorts. I.e., time from target cohort to first event cohort. Time from first to second event cohort (for those with at least 2), time from second to third (for those with at least 3), etc. Distribution of time in days would be useful: mean, sd, p25, median, p75

Actual behavior

Details not available

Steps to reproduce behavior

[pbr6cornell]

I think we should do a proper design session of all the additional statistics we'd like to report out as part of TxPathways. I think @davekern and @chen-regen separately have come up with interesting ideas that should be considered as part of ATLAS 3.0.