Add support for multimodal openai - early version

[fm1320]

The PR adds multimodal (text + image) support to the existing OpenAI client while maintaining backward compatibility with text-only operations. I also adds image generation with Dall E 2 and 3. It also adds tests and updates docstring

1. Unified Input Handling

def convert_inputs_to_api_kwargs(self, input, model_kwargs, model_type):
    # Handles both text-only and image+text inputs in one place
    # Supports both simple text and structured messages

2. Image Processing

def _prepare_image_content(self, image_source, detail="auto"):
    # Supports multiple image input types:
    # - Local files (converts to base64)
    # - URLs (direct use)
    # - Pre-formatted content

Add DALL-E Image Generation Support

3. Added DALL-E 2 & 3 support to OpenAI client for image generation, variation, and editing. Users can now:

• Generate images from text prompts
• Create variations of existing images
• Edit images using masks
• Get results as URLs or base64

Key Changes

• Added IMAGE_GENERATION model type
• Enhanced client with DALL-E API integration
• Added response parsing for image operations
• Maintained existing error handling pattern

Example use:

Text only:

client = OpenAIClient()
response = client.call(
    api_kwargs={"input": "Hello", "model": "gpt-3.5-turbo"}
)

Multimodal:

client = OpenAIClient()
response = client.call(
    api_kwargs={
        "input": "Describe this",
        "model": "gpt-4o",
        "images": "path/to/image.jpg"
    }
)

Image generation:

class ImageGenerator(Generator):
    """Generator subclass for image generation."""
    model_type = ModelType.IMAGE_GENERATION
    
       dalle_gen = ImageGenerator(
        model_client=client,
        model_kwargs={
            "model": "dall-e-3",
            "size": "1024x1024",
            "quality": "standard",
            "n": 1
        }
    )
    
    # For image generation, input_str becomes the prompt
    response = dalle_gen({"input_str": "A happy siamese cat playing with a red ball of yarn"})
    print("\n=== DALL-E Generation ===")
    print(f"Generated Image URL: {response.data}")

TODO:

• Everything shout be an Output Generator type - Generator cant raise error but put the error in error field
• Image generation
• How to raise and catch the error
• parsed chat completion has to be a generator output - inside chat completion parser

Fixes #<issue_number>

Before submitting

• Was this discussed/agreed via a GitHub issue? (not for typos and docs)
• Did you read the contributor guideline?
• Did you make sure your PR does only one thing, instead of bundling different changes together?
• Did you make sure to update the documentation with your changes? (if necessary)
• Did you write any new necessary tests? (not for typos and docs)
• Did you verify new and existing tests pass locally with your changes?
• Did you list all the breaking changes introduced by this pull request?

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[liyin2015]

(1) image output potentially (2) test generator (3) test using real api by yourself