[FEA] Support memory resources from CCCL 3.1.0+

[bdice]

Is your feature request related to a problem? Please describe.
Recently, RAPIDS/CCCL nightlies began failing due to a change in CCCL's memory resources. See NVIDIA/cccl#5313

Describe the solution you'd like
RMM should support these changes in CCCL, which are currently in the main branch (targeting CCCL 3.2.0+) but are also being backported to CCCL 3.1.0+. RAPIDS is currently using CCCL 3.0.2.

[bdice]

Current plan for adoption:

1. Build RMM w/ 3.1 via polyfill and allocate updates (Support building with CCCL 3.1.0 #2017) (25.10)

• Need to verify that all of RAPIDS builds with CCCL 3.1 with these changes in RMM, and ask Spark to do testing with the same pre-release of CCCL 3.1. The goal is to unblock adoption of CCCL 3.1 for RAPIDS.
• Then hopefully CCCL+RAPIDS CI should work.

2. Update polyfill to new allocate signature (RMM internal refactoring) Add CMake option for disabling legacy MR interface #2089 (25.12)
3. Refactor RAPIDS to use new allocate signature (25.12)
4. Adopt any_resource in custom containers (25.12)

• Things storing RMM resource_refs should instead store cuda::mr::any_resource
• Most of these changes will be in RMM, but there are some RAPIDS libraries with custom containers
• Start with device_buffer

5. Adopt any_resource in device-resource global mapping (25.12)
6. Deprecate old allocate signature and pointer-based resources (25.12)
7. Get rid of deprecated code (26.02)

[bdice]

CCCL 3.1.0 + RAPIDS 25.10 CI is working, and RAPIDS is nearly ready to adopt the new CCCL version in 25.12. This will complete task (1) above.