refactor codebase for new euclid workflow usage with tests

[Tooyosi]

Changes for new euclid workflow

What was done:

• Refactor codebase to be less specific to the current cosmic dawn work flow
  • Changes here made on:
  • Label Extractors:
    • finder.rb
    • Create new base_extractor class
    • Create shared cosmic_dawn_and_euclid class
    • Modify decals (the initial workflow configured) extractor class to inherit from the base
    • Modify cosmic_dawn extractor class to inherit from shared class with euclid
    • Create new euclid extractor class
  • Zoobot.rb:
    • Make label_column_headers dynamic
  • create_job:
    • Pass in extractor name to bajor client so it knows what workflow the job belongs to
  • training_data:
    • Fetch the Module(in this case galazy_zoo) and extractor name to generate the proper label_column
  • Schedule.yml:
    • Add new cron job for euclid workflow training
  • bajor/client:
    • pass new options to the request body to match changes made on bajor (pr)

[yuenmichelle1]

Hey Toyosi,

Gonna preface this that I dont really know kade or bajor that well, so this PR review isnt up to snuff of what I usually review. But I think that is ok knowing that the goal is to abstract a bit to get kade working for one other project. And as we find more use cases we can abstract and add more.

I had a couple questions, but also knowing that you and @lcjohnso have been in more discussion about the code changes than I have, if you both to decide to move forward, I'm ok with it .

[yuenmichelle1]

For these migrations, are you planning on backfilling the module_name and extractor_name for any current contexts in production? (Right now there is only 1 context, which if I think about it makes sense, since only 1 project has been using this so far 😆 )

[Tooyosi]

Yes, plan is to use galaxy_zoo and cosmic_dawn or in this new case euclid for the context module_name and extractor_name respectively. We have just one context on the db currently so the aim is to update that after migration

[lcjohnso]

Please do be sure to document and execute this data backfill!