Fix: get_query_results_as_dict integration test

[jtcohen6]

This is a:

• bug fix PR with no breaking changes — please ensure the base branch is master
• new functionality — please ensure the base branch is the latest dev/ branch
• a breaking change — please ensure the base branch is the latest dev/ branch

Description & motivation

See: dbt-labs/spark-utils#7

Unless a query contains order by, its order is not deterministic on analytical databases / data processing technologies. This test is currently failing on Spark even though dbt_utils.get_query_results_as_dict() works just fine.

I believe that order by 1 is supported on all known adapters.

Checklist

• I have verified that these changes work locally on the following warehouses (Note: it's okay if you do not have access to all warehouses, this helps us understand what has been covered)
  • BigQuery
  • Postgres
  • Redshift
  • Snowflake
• I have updated the README.md (if applicable)
• I have added tests & descriptions to my models (and macros if applicable)
• I have added an entry to CHANGELOG.md

[clrcrl]