Generation Failed: User specified an unsupported autocast device_type 'mps'

```
Generation Failed: User specified an unsupported autocast device_type 'mps'
```

This is a blocker. We need to test generation in CI and make sure it actually runs on Apple MPS CPU/GPU
Our app is designed for Apple Metal CPU/GPUs and must fully support them for generation. Lets review the code carefully and fix this.