Rich CK track resolutions avoid transport service

Avoid (slow) calls to the transport service in RichTrackFunctionalCherenkovResolutions by caching the radiation length / unit path length during initialise. In some ways strictly not as accurate, as does not take into account corner cases where tracks clip objects like the beampipe, but in practise this level of precision is not required. In other ways the calculation is now better as previously it would sometimes include terms from objects like the RICH2 mirrors (due to the definition of the start and end points) which really should not have been included, as the parts of the track in the mirrors never give observable photons...

Results in the algorithm being much faster, and means the RICH no longer cares what detector geometry is used (other than for this initial caching).

This MR also extended the monitoring python utilities to allow for different histogram sets, similar to how the old monitoring worked.

This will change the reference logs, so an MR to update them will follow.