Migrate stream APIs from rmm::cuda_stream_view to cuda::stream_ref

[bdice]

Summary

Track the coordinated migration of stream APIs and call sites from rmm::cuda_stream_view to CCCL's cuda::stream_ref. This propagates cuda::stream_ref through RMM containers and memory resources, RAFT resource and handle APIs, downstream C++ interfaces, Python/Cython bindings, benchmarks, tests, and documentation.

This updates RapidsMPF C++ and Python/Cython stream handling to preserve references returned by RMM stream pools and extract raw handles only where required.

Depends on rapidsai/rmm#2372.

Tracked in rapidsai/build-planning#318.

Migrations

• Pass cuda::stream_ref through stream pools, resource accessors, conditionals, and downstream APIs without converting to rmm::cuda_stream_view
• Use cuda::stream_ref constructions for default/legacy/per-thread streams
  • rmm::cuda_stream_default ➡️ cuda::stream_ref{cudaStream_t{cudaStreamDefault}}
  • rmm::cuda_stream_legacy ➡️ cuda::stream_ref{cudaStreamLegacy}
  • rmm::cuda_stream_per_thread ➡️ cuda::stream_ref{cudaStreamPerThread}
• Use .get() when calling an API that requires a raw cudaStream_t, including CUDA runtime, library, CUB, and legacy API boundaries (previously rmm::cuda_stream_view used value())
• Use .sync() when synchronizing a cuda::stream_ref (previously rmm::cuda_stream_view used synchronize())
• Update Cython declarations and call sites to pass stream references directly where supported

[copy-pr-bot]

Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually.

Contributors can view more details about this message here.

[nirandaperera]

LGTM

[pentschev]

Could we update the declaration void wait() declaration to void sync() as part of the migration? I don't think it's currently used anywhere and probably exposed only for completeness, regardless, I think it makes sense to update it (or remove it completely) rather than having a stale one declaration.

[bdice]

Rebased onto preparation PR #1199 and validated against the rebased RMM stack. Local build-rapidsmpf -j0 completed successfully, and the CUDA stream-join tests passed. CI is rerunning at 11a1656d.

[bdice]

CI follow-up: all four devcontainer failures use the pre-migration RMM Cython declaration where Stream.view() returns cuda_stream_view, so Cython rejects forwarding it as stream_ref. The rebased stack includes RMM #2372, where the declaration is migrated; the full build and both CudaStreamJoinCppOnly.* tests pass locally. These failures are expected until the RMM migration is available to CI.