[DOC]: `cuda::maximum` and `cuda::minimum` docs differ from API

[pauleonix]

Is this a duplicate?

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

Is this for new documentation, or an update to existing docs?

Update

Describe the incorrect/future/missing documentation

The docs make it look like these functors take inputs by value, but looking at the header they are taking them by reference instead.

If this is a correction, please provide a link to the incorrect documentation. If this is a new documentation request, please link to where you have looked.

https://nvidia.github.io/cccl/unstable/libcudacxx/extended_api/functional/maximum_minimum.html#libcudacxx-extended-api-functional-maximum-minimum

[fbusato]

also, the documentation should mention the behavior for floating-point types