Harden ROCm detection, fix VRAM heuristic, and expand RDNA2 coverage

- Windows ROCm detection: validate actual GPU presence via hipinfo/amd-smi
  output markers instead of just checking tool existence on PATH
- _ensure_rocm_torch: validate nvidia-smi actually reports a GPU before
  giving NVIDIA precedence (fixes AMD-only hosts with stale NVIDIA tools)
- amd.py _parse_numeric: handle dict-shaped metric objects from newer
  amd-smi versions ({"value": 10, "unit": "W"}) and strip MiB/GiB units
- amd.py VRAM heuristic: raise threshold from 100k to 10M to correctly
  handle MI300X (192 GB = 196608 MB) and other high-VRAM GPUs
- amd.py visible GPU: use AMD-reported GPU IDs instead of enumerate index
  so non-dense sets like CUDA_VISIBLE_DEVICES=1,3 report correctly
- install.sh: add ROCm <6.0 minimum version guard (no PyTorch wheels
  exist for older versions); fix rocm7.1* glob to not match rocm7.10+
- is_rdna: add gfx1033-1036 for RDNA2 mobile GPUs (RX 6600M etc.)
- worker.py: increase ROCm source build timeout from 600s to 1800s;
  fix success log message for ROCm source builds
- Tests: update mocks for _has_usable_nvidia_gpu, add RDNA2 target asserts

studio/install_python_stack.py

@@ -109,6 +109,24 @@ def _has_rocm_gpu() -> bool:
     return False
 
 
+def _has_usable_nvidia_gpu() -> bool:
+    """Return True only when nvidia-smi exists AND reports at least one GPU."""
+    exe = shutil.which("nvidia-smi")
+    if not exe:
+        return False
+    try:
+        result = subprocess.run(
+            [exe, "-L"],
+            stdout=subprocess.PIPE,
+            stderr=subprocess.DEVNULL,
+            text=True,
+            timeout=10,
+        )
+    except Exception:
+        return False
+    return result.returncode == 0 and "GPU " in result.stdout
+
+
 def _ensure_rocm_torch() -> None:
     """Reinstall torch with ROCm wheels when the venv received CPU-only torch.
 
@@ -118,8 +136,8 @@ def _ensure_rocm_torch() -> None:
     (NVIDIA takes precedence).
     Uses pip_install() to respect uv, constraints, and --python targeting.
     """
-    # NVIDIA takes precedence on mixed hosts
-    if shutil.which("nvidia-smi"):
+    # NVIDIA takes precedence on mixed hosts -- but only if an actual GPU is usable
+    if _has_usable_nvidia_gpu():
         return
     rocm_root = os.environ.get("ROCM_PATH") or "/opt/rocm"
     if not os.path.isdir(rocm_root) and not shutil.which("hipcc"):
@@ -707,7 +725,7 @@ def install_python_stack() -> int:
 
     # Windows + AMD GPU: PyTorch does not publish ROCm wheels for Windows.
     # Detect and warn so users know manual steps are needed for GPU training.
-    if IS_WINDOWS and not NO_TORCH:
+    if IS_WINDOWS and not NO_TORCH and not _has_usable_nvidia_gpu():
         if shutil.which("hipinfo") or shutil.which("amd-smi"):
             _safe_print(
                 _dim("  Note:"),