Provide warning on "passing" incremental SQL runs that have production errors

[DylanBaker]

When using the SQL validator in incremental mode, the validator will only test explores that have changed and only show errors that don't exist in production.

Due to how this is done, it can take a while for runs to complete when there are existing production errors on an explore that has been changed. This is because we will (a) find that the explore has changed, (b) run it at the explore level, (c) find that there is an error, (d) run a query for each dimension, and (e) only just to find that it already existed in production.

I don't think there's anything we can do to speed this up, but in case where (1) we've run tests at the dimension level and (2) one of the explores we have done that for is "passing", we should probably show a warning (both in the CLI and the app) that their runs could be significantly sped up by cleaning up the production errors. We could even possibly show the errors.