[FEA]: Host/Device accessors for mdspan

[fbusato]

Is this a duplicate?

• I confirmed there appear to be no duplicate issues for this request and that I agree to the Code of Conduct

Area

CUDA Experimental (cudax)

Is your feature request related to a problem? Please describe.

it is not currently possible to distinguish between host and device accessible mdspan

Describe the solution you'd like

Provide host/device accessors similar to Raft

Describe alternatives you've considered

No response

Additional context

No response

[jrhemstad]

@miscco what are your thoughts on how this would/should interact with #2132? Do we need distinct accessors in addition to the property system?

[miscco]

I was currently thinking we should template the accessors based on the properties of the original container and therewith restrict access to the right memory space