[C2] Implement genAI image and video generation

[ywarnier]

Through #7281 and PR #7304, we already prepared the ground work for the addition of image and video generation.

• Add an icon (for teachers, the same mdi-robot as in other tools) in the documents tool if a provider gives image or video support
• Add a new page where you select between image and video generation if any provider allows them
• Let the teacher choose a directory where it will be stored
• Let the teacher define the name of the image/video to be generated (this is not passed in the prompt)
• Add a prompt space
• Add a "Generate" button
• Show the result in preview (base64 for images - for video I don't know yet)
• Show the modified prompt (if returned by the API, so the teacher knows what the image was generated from, really, and can re-use that modified prompt for another attempt)
• If the teacher accepts the image, add it to the selected directory with the selected name