Doc Update and fixes

[amitsnow]

Summary

This PR includes fixes on documentation and fixes on multimodal support.

Related Issue(s):

• Closes Pipeline fails when audio data is fetched from a dataset from huggingface #76
• Closes multimodal data when being pushed to HF sends some local path instead of the multimodal data #77

Impacted Features:

• Audio data from hf fails to load in the pipeline
• Multimodal data is not getting pushed to hf in expected format

How to Test

Steps for reviewers to verify functionality:

1. Take any pipeline which contains a audio dataset in the source. Run the pipeline. It should run end to end now.
2. Take any multimodal data pipeline and add a hf sink. The multimodal data in hf sink should be a proper multimodal file rather than a path

Screenshots (if applicable)

N/A

Checklist

• Lint fixes and unit testing done
• End to end task testing
• Documentation updated

[psriramsnc]

LGTM 🚀

[zephyrzilla]

Looks good! 🚀