This pull request makes significant improvements to the MoE (Mixture of
Experts) Python API and documentation, focusing on clarifying and
expanding the Expert Parallel (EP) interface, especially around
quantization, dispatch/combine handles, and overlap configuration. The
changes introduce new data structures, update function signatures, and
improve documentation to better reflect the current and planned
capabilities of the system. Additionally, the base development container
is updated to CUDA 13.0, and minor corrections are made to extension
naming.
This PR bumps the project's C++ language standard from C++17 to C++20
(host, CUDA, and HIP) and removes support for CUDA 11, which cannot
compile C++20. CI matrices are updated to drop CUDA 11.8 and add CUDA
13.0 alongside 12.9.
- C++20 enables newer language features across the codebase.
- C++20 requires CUDA ≥ 12.0, so CUDA 11.8 can no longer be supported.
- Blackwell / sm_100 targets require CUDA ≥ 12.8, so CUDA 13.x is added
to CI coverage.
