llvm

5e11fb97 - Improve ScopedContext nullptr handling and re-enable CUDA multi-GPU test

Improve ScopedContext nullptr handling and re-enable CUDA multi-GPU test - Fix ScopedContext to handle nullptr Device without throwing - Only restore context in destructor if original was non-null - Remove CUDA skip from urEnqueueKernelLaunchIncrementMultiDeviceTest - Remove unnecessary P2P support check (P2P is optimization, not requirement)