Fix: Conditionally import `torch.distributed.fsdp` in `trainer_seq2seq.py` (#44507)
* fix: conditionally import torch.distributed.fsdp in trainer_seq2seq
* fix: sort imports in trainer_seq2seq.py
---------
Co-authored-by: DELUXA <you@example.com>
Co-authored-by: Ferdinand Mom <47445085+3outeille@users.noreply.github.com>