[FEA]: Provide `cuda::device::load` and `cuda::device::store` overloadings for "strong" memory accesses

### Is this a duplicate?

- [x] I confirmed there appear to be no [duplicate issues](https://github.com/NVIDIA/cccl/issues) for this request and that I agree to the [Code of Conduct](CODE_OF_CONDUCT.md)

### Area

libcu++

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

https://github.com/NVIDIA/cccl/pull/4003 covers "weak" memory accesses. The same functions can be extended to support "strong" memory accesses.

### Describe the solution you'd like

Provide `T load(const T*, memory_order, ...)` and `store(T data, T*, memory_order, ...)`

### Describe alternatives you've considered

_No response_

### Additional context

_No response_