Add Whisper as a Supported Model

[BujSet]

🚀 The feature, motivation and pitch

OpenAI's whisper model is a SOTA model for automatic speech recognition. Officially supporting this model via examples and would promote the visibility of Executorch and enable audio applications on edge deivces.

Alternatives

No response

Additional context

No response

RFC (Optional)

No response

[cccclai]

whisper is enabled for qualcomm backend https://github.com/pytorch/executorch/tree/main/examples/qualcomm/oss_scripts/whisper and that can be a reference

[larryliu0820]

@cccclai This is very helpful! I think we should extract this out to be a generic whisper runner for different backends. Looking at the code it's not tied to Qualcomm

[BujSet]

PR is WIP.