[Fixed] Model Null when using Davinci

We should change the model to look at the response instead

The response model has versioned names beyond what the user often asks for and causes confusion. Like the model input is gpt-3.5-turbo and the output is gpt-3.5-turbo-0301 which is confusing.

I see- we should look for engine then at least if model is not there

Ahh that explains the bug, yeah lets do that

This should be fixed here
https://github.com/Helicone/helicone/pull/289

working great!

[Fixed] Model Null when using Davinci