Delete any annotated channelmetadata many to many fields to avoid integrity errors

[rtibbles]

Summary

• Currently included_languages is a many to many field from ChannelMetadata to Languages
• Django implements this via a through table
• When we delete the ChannelMetadata object via SQLAlchemy, this does not get properly deleted, because the schema we are using does not know about the included languages table
• This is because using a schema that did know about it would cause other errors, so we let sleeping dogs lie
• To workaround this, we explicitly delete any many to many through tables that are pointing at ChannelMetadata in order to prevent Integrity errors caused by trying to delete a channel metadata row when a many to many through row still points at it
• Adds a regression test that does not replicate the integrity errors, but does at least assert that the deletion of the included languages has occurred.

Reviewer guidance

Does the test make sense?
Can you think of another way to try to actually get the reported integrity error to trigger in the test?

---

Testing checklist

• Contributor has fully tested the PR manually
• If there are any front-end changes, before/after screenshots are included
• Critical user journeys are covered by Gherkin stories
• Critical and brittle code paths are covered by unit tests

PR process

• PR has the correct target branch and milestone
• PR has 'needs review' or 'work-in-progress' label
• If PR is ready for review, a reviewer has been added. (Don't use 'Assignees')
• If this is an important user-facing change, PR or related issue has a 'changelog' label
• If this includes an internal dependency change, a link to the diff is provided

Reviewer checklist

• Automated test coverage is satisfactory
• PR is fully functional
• PR has been tested for accessibility regressions
• External dependency files were updated if necessary (yarn and pip)
• Documentation is updated
• Contributor is in AUTHORS.md

[bjester]

Code changes make sense, but I haven't tested