Aung/fix text tokens and image tokens not documented or typed

[NikkiAung]

• [x] I understand that this repository is auto-generated and my pull request may not be merged

Changes being requested

In your IDE without definition:

details = response.usage.prompt_tokens_details details.text_ # ← No autocomplete! details.text_tokens # ← Type checker error!

In your IDE with definition:

details = response.usage.prompt_tokens_details details.text_ # ← IDE autocompletes to "text_tokens"! ✅ details.text_tokens # ← Type checker OK! ✅

Additional context & links

Since below code is allowing extra fields, we can still access the image_tokens and text_tokens, but it's not good practice knowing that 'text_tokens' and 'image_tokens' is obviously available inside 'prompt_tokens_details'

class Config(pydantic.BaseConfig):
    extra: Any = pydantic.Extra.allow  # ← This allows extra fields!

To test this, need to use gpt-4o-audio model (gpt-4o-audio-preview) with kaggle audio dataset to make mini audio_to_text.py to be able to output 'text_tokens' and 'image_tokens' inside 'prompt_tokens_details'

Next Step

Updating in one of the example documentations here to include text_tokens and image_token inside usage > prompt_tokens_details

Fixed #2554