update coloc readme to work with R14 data

[Lipastomies]

Imported R14 coloc data, and in the process noticed that some things had been changed.

Added preprocessing step to coloc import, and fixed columns where necessary.

[tobtobtob]

Read this through, looks ok to me. This data loading pipeline is quite complicated. In future we should try to make it simpler, by for example extracting the python scripts into files, so that they can be run directly from the command line. Similarly to what we do in SQL repo. Maybe also this could be moved to SQL repo..? But not to do in this PR, I get that this is just an update to the old pipeline, no need to refactor this now.

(I'm not familiar with the colocalization stuff so maybe it would be good if someone else reviews this also)

[juhaa]

Looks ok to me as well. As mentioned, this is quite complicated and could be simplified by quite a lot but needs time to be done properly. And definitely agree that this should be moved to the SQL repo.

[juhaa]

This was unknown to me. Need to fix this directly in the source rather than here.

[Lipastomies]

Agree with that. We could do the following:

1. Move all sql statements, like the coloc & variant table creation, index & view creation to a create_coloc.sql
2. Make a script/scripts that do the necessary processing for the data to be importable directly into the tables using gcloud sql command.

That way it would be a four-step operation:

1. create sql tables
2. process data to ingestable form
3. copy to bucket
4. import to sql tables