Plan for CMS analysis

[MoAly98]

• Move existing coffea-like code into repository [@MoAly98]
• Collect relevant dataset queries ( link) [@pfackeldey] (done in 8017084)
• Write/use dasGO/rucio script to extract all relevant files from queries [@MoAly98] (done by @pfackeldey in queries.py)
• Fine-tuning and cross-checking [@MoAly98]
• Systematics (done in 0ee402c)
  • Collect JSONs for relevant systematics/corrections [@pfackeldey]
  • Summarise systematics in a table (JSONs and links to twikis [@pfackeldey], relevant keys and handles [@MoAly98])
• Integrate all information in configuration for analysis code [@MoAly98]
• write skimming + analysis code in coffea [both]

[MoAly98]

TODOs::

• fix weight branch propagation from datasets
• fix the file writing to avoid bottlneck (parquet? safetensors?)
• Refactor parts of metadata->skimming code to simplify codebase
• ...

[pfackeldey]

FYI for CMS corrections: https://cms-analysis-corrections.docs.cern.ch/