[TorchFX] `quantize_pt2e`: support QAT mode

[daniil-lyakhov]

🚀 Feature request

The task is to support QAT for quantize_pt2e: for that, one need to introduce quantize_qat_pt2e aligned with prepare_qat_pt2e

Feature Use Case

QAT for TorchFX backend

Are you going to submit a PR?

• Yes I'd like to help by submitting a PR!