fix: updating poetry.lock before local wheels are used in solution template

[JoseRamonRodriguez]

It is considered as a good practice that whenever you change the pyproject.toml file, you should run poetry lock --no-update afterwards to sync the poetry.lock files with those changes. This is what we do when some packages are provided as wheel files through the setup_environment.py. The --no-update flag tries to preserve existing versions of dependencies.

In this PR I am executing poetry lock --no-update as part of setup_environment.py before installing dependencies.

[nick-marnik]

@JoseRamonRodriguez Can you please elaborate on the context that triggered the need for this change?

I am not totally opposed to it, but I'm not sure it is the right approach to take. To prevent situations where the TOML file and the lock file get out-of-sync, we use:

• pre-commit hooks for local development
• CI/CD checks (that should be enforced at the PR level)

Specific to the solution template in this repo, we also rely on the E2E solution tests to validate the lock file since the TOML file contains placeholders and is technically invalid TOML. These tests tend to break and I'm not sure they are monitored, so we may need to investigate their current state.

We also have the -F option available on the setup script to allow developers to delete the poetry.lock file.

My concerns:

• If there are situations where these files are out-of-sync, we should try to understand how that is happening and enforce it at the CI/CD level instead
• I think that there should be manual intervention for a developer to update the poetry.lock file. With this change, a developer could be setting up a clean development environment and end up with local changes that they do not understand.