Enhance VQA Task Support for Huggingface Models

[chakravarthik27]

Description

We want to enhance the Visual Question Answering (VQA) task support for Huggingface models within the langtest repository. This will involve extending the current capabilities to integrate better and leverage Huggingface's extensive model library.

Objectives

• Integrate additional Huggingface VQA models.
• Ensure compatibility with existing VQA tasks within the repository.
• Update documentation to reflect the new enhancements.

Benefits

This enhancement will allow users to leverage a broader range of models for VQA tasks, improving flexibility and performance for various applications.

Additional Context

The langtest repository focuses on delivering safe and effective language models, and this enhancement aligns with our goals of expanding and improving model support.