Enable quantization as default for XNNPack for previous failing models (#6242)

Summary:
Pull Request resolved: https://github.com/pytorch/executorch/pull/6242

Since master has migrated aot_compiler to use to_edge_transform_and_lower in a previous change https://github.com/pytorch/executorch/pull/6026, quantization XNNPack options can be enabled by default for the following models:

- Quantized ViT
- Quantized Mobilebert
- Quantized Emformer Predict
- Quantized Emformer Transcribe

Reviewed By: digantdesai

Differential Revision: D64081319

fbshipit-source-id: 4e8ff77af442dfded043c5a5583466afec6beb4e