#294 [错误] 课程代码仅限 CUDA,不支持 Apple Silicon (MPS) 设备路径

On an Apple Silicon Mac (macOS 12.3+) with PyTorch installed — confirm the GPU is real: python3 -c "import torch; print(torch.backends.mps.is_available())" → True.
Run a GPU lesson, e.g. Python phases/04-computer-vision/04-image-classification/code/main.py.
Observe device resolves to cpu (line 195: device = "cuda" if torch.cuda.is_available() else "cpu") — the available GPU is never used.

Or run Python phases/00-setup-and-tooling/03-gpu-setup-and-cloud/code/gpu_check.py → prints "CUDA available: False" and exits, reporting no GPU despite MPS being available.

File	Line	Problem
`phases/00-setup-and-tooling/01-dev-environment/docs/en.md`	Step 6	GPU snippet only shows `nvidia-smi` + CUDA install + `torch.cuda.is_available()`; no MPS equivalent.
`phases/00-setup-and-tooling/03-gpu-setup-and-cloud/code/gpu_check.py`	14–41	Hard-codes `.to("cuda")` and `torch.cuda.synchronize()`; reports no GPU on Apple Silicon.
`phases/00-setup-and-tooling/01-dev-environment/code/verify.py`	19–20	Only probes `torch.cuda.is_available()` / `get_device_name(0)`.
`phases/03-deep-learning-core/11-intro-to-pytorch/code/pytorch_intro.py`	—	CUDA/CPU only device selection.
`phases/04-computer-vision/04-image-classification/code/main.py`	195	`device = "cuda" if torch.cuda.is_available() else "cpu"`.
…