Fix issues with the jsonstorage table

[BryceStevenWilley]

Tried fixing 3 specific things here:

1. Changed db.Index to Index. The former doesn't work anymore, giving an error AttributeError: 'Engine' object has no attribute 'Index'.¹

2. We switched our jsonstorage table to an externally hosted SQL database recently and have run into issues in production with

   (psycopg2.errors.UniqueViolation) duplicate key value violates unique constraint "jsonstorage_pkey" DETAIL: Key (id)=(...) already exists."

   This error has happened twice within a week. This shouldn't be happening, but from reading the SQLAlchemy docs it seems that we need to use server_default to prevent SQL Alchemy from automatically assuming the default of the value? I'm less sure this fix is correct, but it doesn't change current functionality.

3. When we do run into a call that violates the duplicate key, all other writes to the jsonstorage table will fail with

   This Session's transaction has been rolled back due to a previous exception during flush. To begin a new transaction with this Session, first issue Session.rollback(). Original exception was: (raised as a result of Query-invoked autoflush; consider using a session.no_autoflush block if this flush is occurring prematurely) (psycopg2.errors.UniqueViolation) ...

   To prevent this from happening, this change catches excepts that occur when writing or deleting from the jsonstorage table and calls rollback().

Footnotes

1. I can't find where this is failing during normal docassemble runs, but the error can be recreated in python. ↩