[FEA]: Support disjoint temporal neighbor sampling

[haoweiyu-kumo]

Is this a new feature, an improvement, or a change to existing functionality?

New Feature

How would you describe the priority of this feature request

Critical (currently preventing usage)

Please provide a clear description of problem this feature solves

Add support disjoint temporal neighbor sampling.

Currently, there is a hard failure in

cugraph/cpp/src/sampling/temporal_sampling_impl.cuh

Line 118 in 243d875

"Invalid input argument: disjoint sampling is not supported for temporal neighbor "

Describe your ideal solution

Support disjoint temporal neighbor sampling.

Describe any alternatives you have considered

No response

Additional context

No response

Code of Conduct

• I agree to follow cuGraph's Code of Conduct
• I have searched the open feature requests and have found no duplicates for this feature request

[ChuckHastings]

Temporal sampling has been changed to require disjoint as part of the time window sampling PR #5603). During the discussion we determined it didn't make sense to have temporal sampling not be disjoint (at least our current implementation which enforces increasing or decreasing times), since we would have ambiguity to which time to use.

We can discuss as part of #5593 whether to remove the requirement for disjoint sampling if we simply use the time windows alone (without enforcing times increasing or decreasing as we sample).