Fix quantized_matmul with 4D inputs (#4335)

Summary:
Pull Request resolved: https://github.com/pytorch/executorch/pull/4335

MobileBERT has a matmul with 4D inputs (`[1, 4, 8, 32]` by `[1, 4, 32, 8]`) which is erroring out AoT in the meta kernel.

This diff fixes the meta kernel to handle cases where the leading dimensions are more than one (the kernel itself can already handle it!).

Also move the exported graph dump to `export_program`, where it belongs. This prevents some double printing in some cases.

Note: this diff needs at GH approval!

Reviewed By: dulinriley, zonglinpengmeta

Differential Revision: D60050087

fbshipit-source-id: de09ed2fb9c5cdf729cc020119bf090d0f0c70c4