[Paged-Attention] Handle continuous batching for repetition penalty (#39457)
* Handle continuous batching for repetition penalty
* fix last scores and with token mask creation
* add test
* Update src/transformers/generation/continuous_batching.py
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>
* Update src/transformers/generation/logits_process.py
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>
* fix formatting
* remove unneeded cast
---------
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>