User database that persists in their repository

[Rmolimock]

The Problem

1. We can't give users access to our database for security and payment reasons.
2. When their project gets deployed on our site, they will need access to a database, but we don't want to be responsible for paying for all that (potentially infinite) data or have that kind of liability.
3. We don't want them to have to figure out how to set up a remote connection to a database just to make their deployed project work.

Potential Solution
They can store their data in files inside their repository. Then the data they use during development is the same that gets cloned into our containers.
I'm still not sure how we limit their data but this is a step in the right direction.

The Hypothesis
Test the use of SQLite in the cloud run dummy repository. See if you can add and read data both locally and in the deployment container (not the tester container, the one that stays running). As usual don't worry about making it scalable, we just want to prove that this is the way to solve the issue. Try to save some hard-coded string.

New Problem
Cloud Run may not allow read/write of files - fingers crossed.
If it does, we'll have to push to their repository on their behalf. When new data is saved (or a certain amount of it) we will want to add the sqlite file to github in case the container has to be restarted or the repo cloned again for some reason.

If you have any thoughts on the new problems comment on this issue.