Kafka integration

[geoffreyaldebert]

• [x] Kafka Integration (only consumer)
• [x] Read message from udata-analysis-service
• [x] Parse file (could be from minio instead of downloading again resource)
• [x] Add csv-detective type detection to help agate to store resource into sqlite
• [x] Add pandas profiling analysis (minimal) and generation of json report
• [x] Store new infos into sqlite in new tables :
  • general_infos : basic info on resource
  • column_infos : basic info on each column of resource
  • categorical_infos : categorical values for each columns (limit to 10)
  • top_infos : top values for each columns (limit to 10)
  • numeric_infos : basic info on each numeric column of resource (mean, std, min, max)
  • numeric_plot_infos : repartition of values of numeric column in a plot
• [x] Update API to list those new info if we have them

[abulte]

This branch is now published on pypi https://app.circleci.com/pipelines/github/etalab/csvapi/91/workflows/09dba6e2-b91f-4cf2-af03-71a9daee9bbb/jobs/605

⚠️ remove this publication when merged on master