Fix deprecation warnings: torch.load weights_only, cupy fromDlpack, as_target_tokenizer

fix: add Union to typing imports in cl_collator

fix: add Optional/Any import to cl_collator, ipdb try/except in score.py

fix: LoRA reinit, gradient checkpointing use_reentrant=False, checkpoint existence check, collator padding alignment

add root_gainlora to repo for testing

reduce rubish

update

clean code

revert: discard all 2-GPU DataParallel/DDP changes, back to f51f791

DDP 2 GPU

test root

new change