Train adapters on top of FP8 weights (#50031)
* Pass the input gradient through FP8 layers, so adapters can train on top of them
* Same backward for FP8GroupedLinear, with a contiguous gradient for the grouped matmul
* Dequantize in place in the FP8 backward
* Fit the EP sentinel comment in five lines
---------
Co-authored-by: Cyril Vallez <cyril.vallez@huggingface.co>
Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com>