datasets 2.5.0, CMS plotting code

[jpata]

CMS 2.5.0:

• truth jets from CMSSW here are defined as ak4GenJet, not ak4GenJetsNoNu, therefore they include unreconstructable neutrinos and bias the final plots
  • therefore, redefine truth jets in postprocessing2.py to not include neutrinos
• rename gp_ to cp_ for consistency in postprocessing2.py
• split tfds generation to independent subsets 1-10, such that they can be generated separately
• plotting code for CMS dataset validation using high statistics (mlpf/data/cms/plot_cms.py)

CLIC 2.5.0:

• split tfds generation to independent subsets 1-10, such that they can be generated separately
• add more stats for qq and ww_fullhad

TODO:

• regenerate postprocessing
• regenerate tfds
• retrain models
• upload model evals: https://huggingface.co/jpata/particleflow/blob/main/cms/v2.1.0/pyg-cms_20241101_090645_682892/README.md