Fix CUDA build: use PyTorch CUDA index URL explicitly

Default torch on PyPI is CPU-only on Windows. Must use PyTorch's own
package index (cu126) to get CUDA-enabled wheels. This also pins the
CUDA version on Linux for deterministic builds.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

python/build_sidecar.py

@@ -119,8 +119,11 @@ def create_venv_and_install(cpu_only: bool) -> Path:
             "--index-url", "https://download.pytorch.org/whl/cpu",
         )
     else:
-        print("[build] Installing PyTorch (default, may include CUDA)")
-        pip_install("torch", "torchaudio")
+        print("[build] Installing PyTorch (CUDA 12.6)")
+        pip_install(
+            "torch", "torchaudio",
+            "--index-url", "https://download.pytorch.org/whl/cu126",
+        )
 
     # Install project and dev deps (includes pyinstaller)
     print("[build] Installing project dependencies")