Remove unused rows_affected column from test_executions

[jaypeedevlin]

This looks like it may have been copied over from model_executions in error.

Removing it is a non-trivial task because of the migration path — our insert statements don't enumerate the column names, so just removing all mentions of the column will cause inserts for existing projects to fail. We'll need to find a way to get only the column names that exist in the current incremental source model and use those.

This will be generically useful any time we need to drop a column, which we're likely to need to do again.

[glsdown]

This should be possible once #381 has been merged, however, I initially included it in there, and found that DBX is having issues when a field gets removed as it gives an error that it's unable to find the field.

Runtime Error in model test_executions (models/sources/test_executions.sql)
Cannot resolve rows_affected in UPDATE clause given columns DBT_INTERNAL_SOURCE.command_invocation_id, DBT_INTERNAL_SOURCE.node_id, DBT_INTERNAL_SOURCE.run_started_at, DBT_INTERNAL_SOURCE.was_full_refresh, DBT_INTERNAL_SOURCE.thread_id, DBT_INTERNAL_SOURCE.status, DBT_INTERNAL_SOURCE.compile_started_at, DBT_INTERNAL_SOURCE.query_completed_at, DBT_INTERNAL_SOURCE.total_node_runtime, DBT_INTERNAL_SOURCE.failures, DBT_INTERNAL_SOURCE.message, DBT_INTERNAL_SOURCE.adapter_response.; line 11 pos 2

When I have some time I will look into why removing the field causes this error (Snowflake and BQ are fine), but recording it here for future.