Refactor db code to work with existing connections

[kgodey]

Problem

In #578, we updated some of the code in the db module to use existing connections so that we could perform multiple operations in a single transaction. This was done in an ad-hoc basis and isn't easy to read or extend.

Proposed solution

We should figure out an idiomatic and clear way to pass connections and tables around between functions and refactor the code accordingly.

This code should also involve refactoring all the code in db/columns.py to use connections so that:

• code for batch altering and altering a single column can be consolidated
• all code for duplicating a column happens in a single connection

Additional context

• See discussion on Update table PATCH endpoint to work with columns #578 for some ideas.
• Schema sqlalchemy engine caching is per-instance #280 is probably related.
• Connection management in the backend #731

[github-actions]

This issue has not been updated in 90 days and is being marked as stale.

[kgodey]

Closing this, too old, requirements are out of date. We can create a new issue if we need this functionality in the future.