[enhancement]: Adding a conversational image generation feature compatible with OpenAI endpoints

### Is there an existing issue for this?

- [x] I have searched the existing issues

### Contact Details

_No response_

### What should this feature add?

Generating images, particularly image-to-image transformations, currently requires complex prompts that aren’t user-friendly. It would be beneficial to add a conversational image generation feature compatible with OpenAI and other endpoints. This feature would dynamically generate appropriate prompts based on user conversation, executing them directly. Ideally, users could initially select a workflow, and when choosing ‘Image to Image,’ they'd simply upload an image and input something like “make this look anime.” The system prompt generation process itself could be leveraged – by incorporating it into standard openwebUI or similar interfaces with a simple system prompt, we could achieve the same result without needing to integrate with invokeAI. However, this does require you to re-enter the prompt that is displayed, which is a hassle.

### Alternatives

_No response_

### Additional Content

_No response_