Point testing dependencies to main

[mikealfare]

All testing dependencies should point to main to ensure that the latest changes are reflected.

• Move all optional dependencies into hatch environments
  • They aren't production dependencies
  • We can't install from GitHub in optional dependencies and also publish to PyPI
• Make the default environment the dev version of all production dependencies
• Inherit from the default environment for non-detached hatch environments (typechecking and testing)
• Pin lower bound for dependencies in alignment with our practices

Checklist

• I have read the contributing guide and understand what's expected of me
• I have run this code in development and it appears to resolve the stated issue
• This PR includes tests, or tests are not required/relevant for this PR
• This PR has no interface changes (e.g. macros, cli, logs, json artifacts, config files, adapter interface, etc) or this PR has already received feedback and approval from Product or DX

[mikealfare]

The test cases added here are being migrated back to dbt-postgres (for now). It was determined that the issue was likely being caused by a combination of not looking at main for internal dependencies and not instantiating the invocation context variables before checking them. set_invocation_context({}) is run on line 338 now, right before getting the catalog, and tests are now passing.

[mikealfare]

Dependencies were moved down into the hatch environments, in alignment with dbt-adapters. This was done for a combination of reasons:

• we cannot publish a package to PyPI with dependencies installed from GitHub, so development versions of internal packages would need to be identified as hatch environment dependencies
• there is not a lot of value in installing dev and test tooling without the development versions of internal packages
• having just the tooling as installable extras in the production distribution is confusing

[colin-rogers-dbt]

kinda surprised we don't need dbt-tests-adapter for unit tests