Added logging for content filter

[dtee1]

• [ ] I have added tests that cover my changes.
• [ ] If adding a new instrumentation or changing an existing one, I've added screenshots from some observability platform showing the change.
• [ ] PR name follows conventional commits format: feat(instrumentation): ... or fix(instrumentation): ....
• [ ] (If applicable) I have updated the documentation accordingly.

[CLAassistant]

All committers have signed the CLA.

[nirga]

Thanks @dtee1 but I think we already wrap everything with a try-catch and log an exception (see the @dont_throw annotation)

[dtee1]

Thanks @dtee1 but I think we already wrap everything with a try-catch and log an exception (see the @dont_throw annotation)

Ok sounds good. I can take that out. Would you rather I just log the warning in an else?

[nirga]

Thanks @dtee1 but I think we already wrap everything with a try-catch and log an exception (see the @dont_throw annotation)

Ok sounds good. I can take that out. Would you rather I just log the warning in an else?

We also log there 😀

[dtee1]

Thanks @dtee1 but I think we already wrap everything with a try-catch and log an exception (see the @dont_throw annotation)

Ok sounds good. I can take that out. Would you rather I just log the warning in an else?

We also log there 😀

Ok, how is the fix #855 supposed to be implemented? Maybe I missed something

[nirga]

@dtee1 so when calling azure it may filter the response returned from the model if it contains harmful text, see https://learn.microsoft.com/en-us/azure/ai-services/openai/concepts/content-filter?tabs=warning%2Cpython-new

We should add attributes to the span according to those returned from azure in this case.

[nirga]

Superseded by #1539