CMNoise + small updates to other calibrations
This MR includes
- Rewritten CMNoise measurement
- measure number of hits at user-specified threshold at chip, hybrid, and module levels
- 2D plot of correlation between the channels at chip, hybrid, and module levels
- fitting of CM noise at chip level
- metadata tree in output files (currently just includes username and chip IDs)
- Tweaks to
PedeNoise::Validate
Possible improvements to the CMNoise calibration include:
- masking noisy channels
- reducing the side of the 2D histograms -- currently 1 bin per channel and results in a quite large for the module-level plots. Making the 2D histograms is currently optional
- fitting CM noise at larger level -- the fit function requires computing a binomial coefficient, which involves computing
[channel number]
factorial, and doing so overflows a double starting at the hybrid level.