[MPS] Add support for aten::unique_consecutive for MPS backend #88487
Description
🐛 Describe the bug
classifier = pipeline("zero-shot-classification",
model="facebook/bart-large-mnli", device = torch.device("mps"))
candidate_labels = ['label1', 'label2', 'label3'] #list of labels to tag
classifier("SAMPLE TEXT INPUT" , candidate_labels)NotImplementedError: The operator 'aten::unique_consecutive' is not current implemented for the MPS device. If you want this op to be added in priority during the prototype phase of this feature, please comment on https://github.com/pytorch/pytorch/issues/77764. As a temporary fix, you can set the environment variable `PYTORCH_ENABLE_MPS_FALLBACK=1` to use the CPU as a fallback for this op. WARNING: this will be slower than running natively on MPS.
Versions
NA
cc @kulinseth @albanD @malfet @DenisVieriu97 @razarmehr @abhudev