feat: support NCCL AllReduce

[Ziminli]

Summary

This PR adds NVIDIA NCCL support for AllReduce, including the public unique-ID and rank-based communicator initialization APIs needed for NCCL-style bootstrap flows. It also teaches the generated C bridge to return infinicclNotSupported for unavailable operations, adds backend/device combination filtering, includes CCL-only plus OpenMPI-assisted NCCL example programs, and documents NCCL as a partial backend option.

Changes

• Public API and Dispatch

  • Add infinicclUniqueId, infinicclGetUniqueId(), and infinicclCommInitRank() to the public communication API.
  • Add infinicclNotSupported / ReturnStatus::kNotSupported.
  • Add IsSupportedCombination so unsupported backend/device pairs do not instantiate invalid operation implementations.
  • Update bridge generation so operations missing from the selected backend configuration return NotSupported instead of failing to compile.

• NCCL Backend

  • Add NCCL checks, communicator instance handling, data type mapping, and reduction-op mapping.
  • Add NCCL implementations for GetUniqueId, CommInitRank, CommDestroy, and AllReduce.
  • Add runtime GetDevice aliases needed by rank-based communicator initialization.

• Hybrid Communicator Ownership

  • Allow CommInitAll and CommInitRank to populate separate inter_comm and intra_comm slots on one Communicator.
  • Add RAII cleanup for OpenMPI and NCCL communicator instances so mixed communicators release all backend handles on destroy.

• Examples and Runner Tooling

  • Add a thread-per-GPU single-node CCL AllReduce example.
  • Add an OpenMPI-assisted NCCL AllReduce example that uses MPI for rank discovery and unique-ID broadcast, then initializes an NCCL communicator.
  • Update scripts/run_examples.py to forward executable arguments after --.

• Documentation and PR Template

  • Add NCCL to the README supported-backends table as a partial NVIDIA-only backend.
  • Add NCCL to the PR template affected/tested backend checklists.

Platform and Backend Affected

Platform

• CPU
• NVIDIA GPU
• Iluvatar GPU
• MetaX GPU
• Moore Threads GPU
• Cambricon MLU

Backend

• OpenMPI
• MPICH
• NCCL

Performance Impact

• No performance impact
• Performance improved
• Performance regression possible

This adds a GPU-native NCCL path for NVIDIA AllReduce, avoiding the existing MPI host-staging path when the NCCL backend is available. Existing MPI collectives are intended to remain unchanged.

Known Issues & Future Work

• The NCCL backend added here currently covers GetUniqueId, CommInitRank, CommDestroy, and AllReduce; other NCCL collective operations remain future work.

Test Results

Test Involved Platform

• CPU
• NVIDIA GPU
• Iluvatar GPU
• MetaX GPU
• Moore Threads GPU
• Cambricon MLU

Test Involved Backend

• OpenMPI
• MPICH
• NCCL

Pure CCL (NCCL) on single-node NVIDIA:
ccl_all_reduce.log

CCL + MPI on single-node NVIDIA:
ccl_mpi_hybrid_all_reduce.log

MPI on Heterogeneous Cluster:
mpi_all_gather.log
mpi_all_reduce.log
mpi_all_to_all.log
mpi_broadcast.log
mpi_gather.log
mpi_reduce.log
mpi_reduce_scatter.log
mpi_scatter.log
mpi_send_recv.log

---

Checklist

Every contributor must verify every item below before requesting
review. Tick each box only after the check has actually been performed —
do not tick speculatively. If an item truly does not apply, replace the
checkbox with N/A and briefly explain why in an inline comment.

Title, Branch, and Commits

• PR title follows Conventional Commits (e.g. feat: …, fix(nccl): …).
• Branch name follows <type>/xxx-yyyy-zzzz where <type> matches the PR title's Conventional Commits type and words are joined with hyphens (see CONTRIBUTING.md §Branches).
• Each commit message follows Conventional Commits.
• Small PR is a single squashable commit; or, for a large PR, every commit is meaningful, well-formed, and independently reviewable (see CONTRIBUTING.md §Pull Requests).
• No stray merge commits from master — the branch is rebased cleanly on top of the current master.
• No fixup! / squash! / wip commits remain.

Scope and Design

• Changes are minimal — no unrelated modifications were introduced (CONTRIBUTING.md §Code/General).
• No dead code, commented-out blocks, debug prints, printf/std::cout/print(...) left behind, or TODO without an owner and issue link.
• No unrelated formatting churn that would obscure the diff.
• Public API changes (if any) are intentional, documented, and reflected in affected callers/tests.

General Code Hygiene

• The code is self-explanatory; comments were added only where the intent or rationale is non-obvious (CONTRIBUTING.md §Code/General).
• Every modified or added file ends with a single trailing newline (CONTRIBUTING.md §Code/General).
• No trailing whitespace, inconsistent indentation, or mixed formatting styles remain.
• Identifiers referenced in comments or error messages are wrapped in Markdown backticks (e.g. the `AllReduce` implementation) (CONTRIBUTING.md §Code/General).
• All comments and error messages are in English (CONTRIBUTING.md §Code/General).
• Comments and error messages are complete sentences — capitalized first letter, terminal punctuation — unless the language/framework convention says otherwise (CONTRIBUTING.md §Code/General; §Python).

C++ Specific (if C++ files changed)

• Code follows the Google C++ Style Guide strictly.
• clang-format (version 16, per .github/workflows/clang-format.yml) has been run against all modified applicable files; the diff is clean.
• No exceptions are thrown. Error paths use assert with messages that include at least __FILE__, __LINE__, and __func__ (CONTRIBUTING.md §C++).
• Error and warning message wording follows the LLVM Coding Standards (CONTRIBUTING.md §C++).
• Constructor initializer list order matches member declaration order (CONTRIBUTING.md §C++).
• Exactly one blank line between classes, between classes and functions, and between functions (CONTRIBUTING.md §C++).
• Exactly one blank line between members (functions and variables) within a class (CONTRIBUTING.md §C++).
• Exactly one blank line before and after the contents of a namespace (CONTRIBUTING.md §C++).

Python Specific (if Python files changed)

• Code is PEP 8 compliant; ruff check passes cleanly on CI (see .github/workflows/ruff.yml).
• ruff format --check passes cleanly — if not, run ruff format and commit the result.
• Comments are complete English sentences, starting with a capital letter and ending with punctuation; Markdown backticks are used for code references (CONTRIBUTING.md §Python).
• Framework-specific conventions (e.g. lowercase pytest.skip messages without terminal period) are honored where applicable (CONTRIBUTING.md §Python).
• No blank line between the function signature and the body when there is no docstring or comment (CONTRIBUTING.md §Python).
• A blank line is present before and after if, for, and similar control-flow statements (CONTRIBUTING.md §Python).
• A blank line appears before each return, except when it directly follows a control-flow statement (CONTRIBUTING.md §Python).
• Docstrings (if any) follow PEP 257 (CONTRIBUTING.md §Python).
• Type hints are added / kept consistent with the surrounding code.

Testing

• All applicable example programs have been built and tested successfully on at least one supported heterogeneous cluster setup.

Build, CI, and Tooling

• New backends or devices have been added to auto-detection in CMakeLists.txt under if(AUTO_DETECT_DEVICES) or to if(AUTO_DETECT_BACKENDS) if applicable.
• Both CI workflows (clang-format.yml, ruff.yml) are green locally (or expected to be green on CI).

Documentation

• README.md, CONTRIBUTING.md, or inline docs updated when behavior, build flags, or developer workflow changed.
• Any user-visible breaking change is called out explicitly under "Summary" and in the commit/PR title with a ! or BREAKING CHANGE: footer.

Security and Safety

• No secrets, access tokens, internal URLs, customer data, or personal hardware identifiers have been committed.
• N/A- Third-party code is license-compatible and attributed.
• No unsafe pointer arithmetic, uninitialized reads, or missing bounds checks were introduced.