Sets up metadata db for every llm class

[PranavPuranik]

Description

The metadata database for storing chat history is a sqllite3 db. It is setup up in the app class by using setup_engine.

But running individual tests in llm folder fails because we don't create any app for tests. Individual llm classes are called independently. Thus, setup_engine is never called.

This change moves the setup to BaseLlmConfig, which is called every time we load an llm.

Fixes #1400

Type of change

Please delete options that are not relevant.

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Refactor (does not change functionality, e.g. code style improvements, linting)
• Documentation update

How Has This Been Tested?

Tested individual tests in llm folder, and also tested running my demo app.

• Unit Test
• Test Script (please provide)

Checklist:

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes
• Any dependent changes have been merged and published in downstream modules
• I have checked my code and corrected any misspellings

Maintainer Checklist

• closes #xxxx (Replace xxxx with the GitHub issue number)
• Made sure Checks passed

[Dev-Khant]

Hi @PranavPuranik Can you please check why tests are failing?

[Dev-Khant]

#1449 will fix psutil not found issue.

[PranavPuranik]

Nice, I will wait till its merged in. All tests are passing my laptop - for every commit. Funny how they end up failing here in CI... Investigating

[Dev-Khant]

Hey, what is the need of adding this test in LLM folder?

[PranavPuranik]

Original Problem: We couldn't run LLM tests individually.

Initially, I had the Alembic fixture under the original conftest file at tests/conftest.py. Tests were passing individually and overall on my laptop from the command line, but they failed on CI. I suspected it was an integration test failure because the error indicated a database issue.

To address this, I moved the fixture inside the LLM folder since we don't need to perform Alembic upgrades for LLM testing but might need it somewhere else. For LLMs, we only test assert_called_once or similar assertions. Moving it inside reduces the scope of the fixture to only run on these LLM tests.

Additionally, PyCharm now picks up this tests/llm/conftest.py, which is great because it wasn't able to pass individual tests when the fixture was in tests/conftest.py. Although this could be due to some configuration issue.

[deshraj]

@PranavPuranik can you please resolve the conflict here? Changes look good.

[PranavPuranik]

resolving

[CLAassistant]

All committers have signed the CLA.