[Tasks] Fix widget model for image-text-to-text task (#973)

This PR updates the widget model for `image-text-to-text` task to put
`meta-llama/Llama-3.2-11B-Vision-Instruct` instead of
`microsoft/kosmos-2-patch14-224` which is an `image-to-text` model.

Also, I'm not sure but this will maybe also fix the widget displayed in
the `image-text-to-text` task
[page](https://huggingface.co/tasks/image-text-to-text).

![Screenshot 2024-10-18 at 15 17
21](https://github.com/user-attachments/assets/c7f8a020-b7c6-4bd3-982a-e729253e3f92)

Co-authored-by: Lucain <[email protected]>

Author: hanouticelina

packages/tasks/src/tasks/image-text-to-text/data.ts

@@ -87,7 +87,7 @@ const taskData: TaskDataCustom = {
 	],
 	summary:
 		"Image-text-to-text models take in an image and text prompt and output text. These models are also called vision-language models, or VLMs. The difference from image-to-text models is that these models take an additional text input, not restricting the model to certain use cases like image captioning, and may also be trained to accept a conversation as input.",
-	widgetModels: ["microsoft/kosmos-2-patch14-224"],
+	widgetModels: ["meta-llama/Llama-3.2-11B-Vision-Instruct"],
 	youtubeId: "IoGaGfU1CIg",
 };