Towards AGC task v2

[alexander-held]

An updated v2 of the ttbar analysis task will

• include an extended set of systematic uncertainties, e.g. more complete jet uncertainty handling (can address this via correctionlib, see also https://gitlab.cern.ch/cms-nanoAOD/jsonpog-integration)
  • first set of correctionlib-based variations by @Andrew42 -> feat: adopt correctionlib for applying systematic variations #119
• add ML component (factorized training & inference), cc @ekauffma
  • training: feat: cms ttbar training notebook #127
  • inference: feat: add ML inference option to ttbar analysis pipeline #122
  • inference follow-up: Follow-up items to ML extension of analysis #130
• use MC generator weights for samples