Method to allocate CUDA tensors on specific device

[ezyang]

PyTorch CUDA tensor constructors have an undocumented keyword argument device which allows you to specify what GPU device the tensor should be allocated on. Looking at Type in ATen (the documented method for allocating tensors), there does not seem to be any way to specify the device when allocating new tensors this way. There should be!

[prigoyal]

@zdevito I think it will be useful to have this feature available. We stumbled upon one use case today when using ATen outside pytorch