Fix the pileup reweight module from compare.py. Now, it uses the same strategy as flatten.py.
Partial ReReco samples from Run2022 (B,C,D and E) dataset are added.
As it can be checked, the PU reweight now works much better: