Add functionality for fake data in inference server db

[wirthual]

Initial code to add fake data for inference db, tracked by #3090

For simplicity, I used Polyfactory to generate the data.

I faked data for: User, Worker, Chat and Message

The message content is loaded from realistic_seed_data.json

I tested the code from a clean Postgres instance locally. The loading of the data was successful.

Looking forward to any feedback regarding this implementation, needed requirements and testing procedures.

[github-actions[bot]]

:x: pre-commit failed. Please run pre-commit run --all-files locally and commit the changes. Find more information in the repository's CONTRIBUTING.md

[github-actions[bot]]

:x: pre-commit failed. Please run pre-commit run --all-files locally and commit the changes. Find more information in the repository's CONTRIBUTING.md

[wirthual]

@olliestanley I changed to code based on your feedback. Thank you for that.

Anything else we need to address?

I set this PR to ready.

[olliestanley]

Sorry for slow review on this - I think this is looking good, have just left a couple of comments

[olliestanley]

Just realised this will need review from Andreas/Yannic also as it touches the docker/ directory. Hopefully one of them will be able to review soon

[revenge47]

@wirthual Sorry technical issues again. The code looks perfect. I had problems creating new sessions for adding the messages. Why did you choose to use Polyfactory instead of just inserting them without it? I'm seeing its compatibility with pydantic as a benefit but I'm assuming doing without it would be the same?

[wirthual]

Yes, you are right. It could also be filled with data stored in a Json file.

The reason I chose polyfactory it made is just easier to fill the database with fake data. Plus if additional fields are added in future, it should just work out of the box.