Need to handle 0 changepoints better #4
Description
Found when running break.R in RK downloads.
The influence measure finds 0 changepoints and handles well but the stability.overview when colouring the changepoints throws an error when there are no changepoints. Same error for location.stability.
A different error in the influence.map but relates to the same problem of 0 changepoints.