Core ML: emit a (1,520,520) int32 index map instead of (1,21,520,520) logits

Core ML: emit a (1,520,520) int32 index map instead of (1,21,520,520) logits

Stop restating quantized, and drop the unused default

Core ML: drop the fp32 variant, fp16 is equivalent in task terms

Add backend config (ExecuTorch 1.4.1)

Add coreml export (ExecuTorch 1.4.1)

Add coreml export (ExecuTorch 1.4.1)