Fix vp-setupfit and vp-rebuild-data
Created by: wilsonmr
In the data keyword PR the function _filter_closure_data
in filters.py
got changed such that the data never got saved
vp-rebuild-data
was broken at some point by a reshuffle of n3fit
which wasn't propagated through.
I have fixed both of these issues and then added a test which filters some data and then rebuilds it and checks that the central value is the same
I then realised that using collect in vp-setupfit was a bad idea because it doesn't guarantee the same result each time, because it can shuffle the ordering of filtering the data so I changed the collect into a simple function with a for loop. This was required to have a consistent test - the proof will be in the pudding (whether the test passes on CI)