Bump CCCL version to include cuda::std::span fix

[sleeepyjack]

Description

This PR updates the CCCL version to include a fix for cuda::std::span which is required for cuCollections to work properly with CCCL 2.5.0.

Most of the changes between the last CCCL version bump (#607) and this one were related to doc updates and unit test fixes, so I don't expect much functional impact for RAPIDS.

After this PR we likely have to bump the cuco version again to include the new changes.

CCCL PR:

• Fix span for non-ranges NVIDIA/cccl#1836

CUCO PR:

• Bump rapids-cmake version NVIDIA/cuCollections#502

RAPIDS PRs:

• [DO NOT MERGE] Test CCCL version bump cudf#15986
  • error during docs-build but seems unrelated
• [DO NOT MERGE] Test CCCL version bump cugraph#4483
  • some CI jobs ran into a network timeout -> rerunning
• [DO NOT MERGE] Test CCCL version bump cuml#5924
  • successful apart from some optional conda python tests
• [DO NOT MERGE] Test CCCL version bump raft#2358
  • This one is weird as it shows an error that is similar to the one we found in cuco when bumping the CCCL version to 2.5.0. However, we thought the problem was on cuco's end and pushed a fix that resolved the issue (see Slack thread)
• [DO NOT MERGE] Test custom rapids-cmake rmm#1584

Checklist

• I am familiar with the Contributing Guidelines.
• New or existing tests cover these changes.
• The documentation is up to date with these changes.
• The cmake-format.json is up to date with these changes.
• I have added new files under rapids-cmake/
  • I have added include guards (include_guard(GLOBAL))
  • I have added the associated docs/ rst file and update the api.rst

[bdice]

The diff is here: NVIDIA/cccl@fde1cf7...e21d607

I feel confident in merging this as soon as you think it's sufficiently tested @sleeepyjack @PointKernel. All of the commits up to those from yesterday (which unfortunately includes the fix you need) were tested by the latest run of NVIDIA/cccl#1667.

[PointKernel]

Added rmm PR as well: rapidsai/rmm#1584 since the raft failure (rapidsai/raft#2358) pointed to an rmm invocation:

Thread 1 "CORE_TEST" hit Catchpoint 1 (exception thrown), 0x00007fffb24824a1 in __cxa_throw () from /usr/lib/x86_64-linux-gnu/libstdc++.so.6
(cuda-gdb) bt 5
#0  0x00007fffb24824a1 in __cxa_throw () from /usr/lib/x86_64-linux-gnu/libstdc++.so.6
#1  0x0000555555654a0a in rmm::mr::limiting_resource_adaptor<rmm::mr::device_memory_resource>::do_allocate(unsigned long, rmm::cuda_stream_view) ()
#2  0x00005555556c3d8f in void* cuda::mr::__4::_Resource_vtable_builder::_Alloc_async<rmm::mr::limiting_resource_adaptor<rmm::mr::device_memory_resource> >(void*, unsigned long, unsigned long, cuda::__4::stream_ref) ()
#3  0x00005555556becd1 in raft::Raft_WorkspaceResource_Test::TestBody() ()
#4  0x00005555557cfee1 in void testing::internal::HandleExceptionsInMethodIfSupported<testing::Test, void>(testing::Test*, void (testing::Test::*)(), char const*) ()

[sleeepyjack]

cugraph CI has some problems. I restarted it one more time. Fingers crossed. Apart from that all other projects seem to be fine with the change.

[sleeepyjack]

cugraph now finally also passes all unit tests. What's going on with the failing rapids-cmake tests? I can't make sense of them and thus would like a second pair of eyes to verify if it's related to this change.

[sleeepyjack]

All problems have been resolved. This PR is ready for review.

[bdice]

/merge

[bdice]

Thanks @sleeepyjack!