The Makefile should test for nvidia-smi, not nvcc, to guess if CUDA available

[mhucka]

Describe the issue

It seems that checking for nvidia-smi is the more reliable way to determine if a system has a usable NVIDIA GPU and the necessary drivers installed for running CUDA applications. Checking for nvcc may only tell you if the system can compile CUDA code, not necessarily if it can run it.

Tell us how to reproduce the issue

No response

Tell us the version of qsim or qsimcirq (if relevant)

No response

Tell us the computing environment (if relevant)

No response

[mhucka]

Testing only for nvidia-smi also is not necessarily indicative of having a GPU. Closing this issue as something that is probably not worth changing.