Describe the bug

Using equal_rowcount between 2 different tables with the same groupby columns results in a false positive test

Steps to reproduce

1. Create 2 different tables, with the same group by column. Let's call this gb. gb can have different values between the 2 tables (but same type). For example: 1 table could be (select 'x' as gb union all (select 'y' as gb)) and the other could be (select 'z' as gb)
2. Apply equal_rowcount between the 2 table using gb as a group_by_columns
3. Run the dbt data test

Expected results

This test should fail and users should be alerted

Actual results

This test passes without any problem

Screenshots and log output

Result of dbt_internal_test if using above example

Because we have coalesce functions in the test evaluation, all null diff_count result as 0 failure, while it should be something else

select
      sum(coalesce(diff_count, 0)) as failures,
      sum(coalesce(diff_count, 0)) != 0 as should_warn,
      sum(coalesce(diff_count, 0)) != 0 as should_error
    from dbt_internal_test

System information

The contents of your packages.yml file:
packages:

• package: dbt-labs/dbt_utils
  version: 1.3.0

Which database are you using dbt with?

• postgres
• redshift
• bigquery
• snowflake
• other (specify: ____________)

The output of dbt --version:
I'm using dbt cloud so I suppose it's versionless

Additional context

I think it's line 63 of equal_rowcount.sql. Instead of abs(count_a - count_b) as diff_count it should be abs(coalesce(count_a, 0) - coalesce(count_b, 0))

If this is applied, diff_count would no longer be null and coalesce() is no longer needed inside the sum() functions above. It could be moved to outside the sum() function like #974 which resolves #973

Are you interested in contributing the fix?

Yes I can contribute