Improve dryrun skip logic for stage deploys

[Anna Scholtz]

Using wildcards to skip dryruns for everything in a project results in all the stage deployed tables being skipped as well due to the project name simply being replaced by the stage env project:

bigquery-etl/bigquery_etl/dryrun.py

Line 101 in 3e34be8

 # update skip list to include renamed queries in stage. 

This started happening after adding:

bigquery-etl/bqetl_project.yaml

Line 26 in 3e34be8

 - sql/data-observability-dev/**/**/*.sql 

We should revise the way dryruns for queries deployed on stage work.

┆Issue is synchronized with this Jira Task