idea: publish pypi package to provide useful utilities for data science oriented users

This package could do things like:
- read credentials from the same places as collectoss and automatically provide a database connection object that can be used to get to a pandas dataframe (ideally without shipping pandas)
- this could also set the connection search path and other technical details many data scientists likely dont care about
- it could also look for the credentials files in many different spots, like the current Dir, the repo root, your homedir, etc.
- encourage people to use the sample data image (coming soon)
- check whether materialized views have ever been refreshed, or when they were last refreshed (internal support for this coming soon)
- maybe find a way to include the same comments that are in our db schema in some kind of help interface so people can more easily learn more about what they are querying
- the existence of this would act to standardize setting up a notebook (and we could provide a sample notebook cell in a readme)

Open to additional value adds as well