Qualcomm AI Engine Direct - support embedding op (#2057)

Summary:
- support embedding op with int32 index input
- make mobilebert / llama2 be fully delegated
- add requantize passes for mixed precision
- bug fixes

Pull Request resolved: https://github.com/pytorch/executorch/pull/2057

Reviewed By: dbort

Differential Revision: D54348816

Pulled By: cccclai

fbshipit-source-id: ec3c8e87cc879d6f642859231255d5094d78349f