cumlprims_mg: move all development to rapidsai/cuml, stop publishing packages

[jameslamb]

Description

Starting in RAPIDS 26.02, cumlprims_mg has been moved into cuml: NVIDIA/cuml#7585

We can stop publishing separate cumlprims_mg packages and maintaining https://github.com/rapidsai/cumlprims_mg

This tracks that work.

Benefits of this work

Reduces maintenance burden and complexity.

Acceptance Criteria

• 0 lingering unnecessary references to cumlprims_mg in RAPIDS 26.02 and beyond
• rapids-metadata accurately reflects the state of the 25.12 and 26.02 releases

Approach

• create an RSN announcing the change (RSN 56: moving cumlprims_mg into cuml docs#730)
• put a note in the README about the archiving (add note about the repo being archived cumlprims_mg#33)
• stop running nightly builds and tests of cumlprims_mg (Drop cumlprims_mg from nightly builds/tests workflows#102)
• remove from RAPIDS build infrastructure ('rapidsai' GitHub search | 'NVIDIA' GitHub search | 'nv-gha-runners' GitHub search | all of GitHub search)
  • build-infra (https://github.com/rapidsai/build-infra/pull/321)
  • devcontainers (remove cumlprims_mg devcontainers#642)
  • rapids-metadata (remove cumlprims_mg from 26.02 rapids-metadata#69)
  • release-scripts (https://github.com/rapidsai/release-scripts/pull/56)
  • CI runner permissions (private repo)
  • telemetry dashboards (private repo)
  • NVIDIA/cccl devcontainers (Remove cumlprims_mg from RAPIDS workflows/devcontainers NVIDIA/cccl#7082)
  • cuml clang-format configuration (Remove cumlprims from clang-format NVIDIA/cuml#7647)
  • cugraph clang-format configuration (Remove cumlprims from clang-format cugraph#5382)
• inform RAPIDS Ops about the change in the set of published packages (https://github.com/rapidsai/ops/issues/4460)
• delete all of the 26.02 nightly cumlprims_mg packages (https://github.com/rapidsai/ops/issues/4467)
• archive https://github.com/rapidsai/cumlprims_mg (https://github.com/rapidsai/ops/issues/4467)
• mark the RSN complete (RSN 56: mark complete docs#731)

Notes

This issue's format was copied from #208 and #210

[jameslamb]

This is done, thanks everyone!