[ADAP-699] [Regression] dbt-redshift is susceptible to breaking changes via patch releases to redshift_connector

[dataders]

Is this a regression in a recent version of dbt-redshift?

• I believe this is a regression in dbt-redshift functionality
• I have searched the existing issues, and I could not find an existing issue for this regression

Current Behavior

Follow up to #518

https://github.com/aws/amazon-redshift-python-driver, a package upon which this repo heavily relies, does not follow Semantic Versioning, despite using three-integer release version names. Instead there's weekly release where the patch number is incremented.

The conventional, Pythonic approach is to compatibility-pin package dependencies to the minor version (~=x.y.z), which allows our users to take advantage of any new, non-breaking, bug fixes shipped in patch releases after we release. This is not possible with redshift_connector.

Instead, we hard-pin to a specific patch, and only consider incrementing the patch version:

• if there's a user-impacting change in redshift_connector, or
• whenever we are ourselves releasing a new major or minor version

Expected/Previous Behavior

xxx

Steps To Reproduce

xxx

Relevant log output

No response

Environment

- OS:
- Python:
- dbt-core (working version):
- dbt-redshift (working version):
- dbt-core (regression version):
- dbt-redshift (regression version):

Additional Context

No response