[FEA]: Explore PDL in segmented sort

### 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

CUB

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

Small and large sorting kernels are serialized in segmented sort. This leads to suboptimal performance on non-uniform segment size distributions.

### Describe the solution you'd like

We can overlap large and small/medium sorting kernels with PDL. This issue can be closed with estimated performance impact of a feature like that.

### Describe alternatives you've considered

_No response_

### Additional context

_No response_