Improve docs on retries

What's the issue or suggestion?

We noticed that our retry policy for jobs stopped working some time ago. After reading the docs, we added the missing config:

run_retries:
   enabled: true

But it didn't help. It took some time to discover that apart from that, we need to migrate our dagster instance to add a missing db table for this to work:

sqlalchemy.exc.ProgrammingError: (psycopg2.errors.UndefinedTable) relation "kvs" does not exist

It would be great to add this information to the docs on retries.

Additional information

No response

Message from the maintainers

Impacted by this issue? Give it a ! We factor engagement into prioritization.