[ENHANCEMENT]OpenAI autoinstrumenation, track cacheing tokens number

[arizedatngo]

Details: Cacheing feature from OpenAI, we should likely track the cached_tokens

https://platform.openai.com/docs/guides/prompt-caching

"usage": { "prompt_tokens": 2006, "completion_tokens": 300, "total_tokens": 2306, "prompt_tokens_details": { "cached_tokens": 1920 }, "completion_tokens_details": { "reasoning_tokens": 0 } }

[arizedatngo]

Additional thought: Might be nice to have the change reflected in the UI

" it will be more consistent if we show the cached_tokens with other existing token count right?
btw, we have not logged cached_tokens yet, will let you know when we start to log it"