Add AMD ROCm GPU support

Introduces a third hardware mode alongside CUDA and CPU: ROCm (AMD GPU).
AMD GPUs present as CUDA devices under PyTorch ROCm, so the existing GPU
path is reused with minimal changes — the main additions are wheel management,
device detection, and suppressing flash_attn (unsupported on ROCm).

- server/vibevoice_server.py: extend _resolve_device() to recognise 'rocm'
  (auto-detected via torch.version.hip); add _torch_device() helper that maps
  'rocm' → 'cuda' for all PyTorch API calls; apply GPU optimisations for both
  cuda and rocm in _init_model(); always use sdpa on ROCm; propagate
  _torch_device() to _load_voice_presets() map_location.
- server/start.sh: add --rocm flag; sync .venv-rocm with uv sync --no-sources
  then replace torch with the ROCm 6.2 wheel via uv pip install; set
  VIBEPOD_DEVICE=rocm for uvicorn.
- server/pyproject.toml: register pytorch-rocm62 index (explicit); add
  .venv-rocm to ruff excludes.
- package.json: add dev:rocm and dev:server:rocm scripts.
- README.md: document ROCm mode, prerequisites (RX 6000+, ROCm 6.2+, Linux),
  and new commands; expand CUDA vs CPU section to CUDA vs CPU vs ROCm.

https://claude.ai/code/session_0168pSswiaoEf6LEx6UQWfBu

package.json

@@ -6,8 +6,10 @@
     "build": "pnpm --filter vibepod-web build",
     "dev": "bash dev.sh",
     "dev:cpu": "bash dev.sh --cpu",
+    "dev:rocm": "bash dev.sh --rocm",
     "dev:server": "bash server/start.sh",
     "dev:server:cpu": "bash server/start.sh --cpu",
+    "dev:server:rocm": "bash server/start.sh --rocm",
     "dev:web": "pnpm --filter vibepod-web dev",
     "format": "prettier --write . && cd server && uv run ruff format .",
     "format:check": "prettier --check . && cd server && uv run ruff format --check .",