fix(ingest/okta): fix event_loop RuntimeError with nested asyncio

[skrydal]

This PR addresses bug #8636. Proper fix for it depends on intended behavior of the okta connector - if nesting event loops is desired behavior then nest_asyncio should be used.

For more considerations see:
https://stackoverflow.com/questions/46827007/runtimeerror-this-event-loop-is-already-running-in-python

Fixes #8636.

Checklist

• The PR conforms to DataHub's Contributing Guideline (particularly Commit Message Format)
• Links to related issues (if applicable)
• Tests for the changes have been added/updated (if applicable)
• Docs related to the changes have been added/updated (if applicable). If a new feature has been added a Usage Guide has been added for the same.
• For any breaking change/potential downtime/deprecation/big changes an entry has been made in Updating DataHub

[hsheth2]

This feels like a bandaid - it just seems like we're using asyncio incorrectly in this source. Technically, we shouldn't be using run_until_complete because it's blocking, and defeats the point of using asyncio in the first place (not to mention doesn't work if we don't own the event loop).

That said, using nest_asyncio is like a reasonable approach vs doing a big refactoring, so probably fine for now.