Migrate monkeypox from ops to ETL

[Marigold]

We're updating monkeypox data from owid-admin-prod and schedule it with CRON. It's hidden and shouldn't run there, so the best would be migrating it to ETL.

[larsyencken]

From triage discussion:

• How should it get run?

  • The data in Github hasn't updated in 2 months -- is it still running?
  • This is super related to Add cron-like scheduling into the ETL for regular updates #3016, and if we want it in the ETL we should decide the approach there first
  • It's a reasonable candidate for the ETL since it doesn't do scraping
  • It would also be ok to self-update using Github actions, then use the repo as input to the ETL (a github step)

• How should we keep the repo CSV up to date

  • Pablo suggested an external channel, which exists already; the code there publishes to Github as a side-effect
  • We could put these into the DAG as an explicit target

Suggest to look at again after #3016

[larsyencken]

Update from @Marigold :

• It's going well, and it's not blocked on the regular updates discussion
• The initial approach is to have a common pipeline for all frequent updates, run daily in Buildkite

[Marigold]

Mpox explorer is now powered by dataset from ETL. We still have to merge changes that would be committing data to github repo.