Make device properties prettier

[leofang]

Similar to #205 but for Device properties. Currently we return all of them at once, with the C enumerator names as the keys

cuda-python/cuda_core/cuda/core/experimental/_device.py

Lines 82 to 85 in 213baf4

	@property
	def properties(self) -> dict:
	# TODO: pythonize the key names
	return handle_return(cudart.cudaGetDeviceProperties(self._id))

We should perhaps make the returned object a Python class instead of dict, with nice, pythonic attributes.

[leofang]

xref: our C++ counterpart (cudax) NVIDIA/cccl#2084

[ksimpson-work]

Piotr suggested we base the exposure on cudaDeviceAttribtues rather than cudaDeviceProperties to reduce the minor version compatibility considerations. WDYT @leofang

[leofang]

Inside you DeviceProperties class we could consider using that. But I don't know how does this help with

to reduce the minor version compatibility considerations

[ksimpson-work]

I guess if cuda 13.x.y came out and .y had some great new property we wanted to expose, we would have to wait for 13.z if we used the property struct in the backend, whereas we could expose it right away if we used the attributes. I will look into it and wee what the trade-offs are