[FEA]: Implement CUDA backend for parallel `cuda::std::set_difference`

### 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.

We want to implement a CUDA backend for the following algorithms:

* `std::set_difference` see https://en.cppreference.com/w/cpp/algorithm/set_difference.html
* `std::set_symmetric_difference` see https://en.cppreference.com/w/cpp/algorithm/set_symmetric_difference.html

### Describe the solution you'd like

Not clear how to implement this.

It is possible that we skip this

### Describe alternatives you've considered

_No response_

### Additional context

_No response_