feat(megamoe): add nvfp4 group16 capability gate

Allow SM100 FP4 scale layout transforms to accept group16 and thread weight granularity through the MegaMoE Python wrapper, API checks, and synthetic benchmark entrypoint.

Keep fused SM100 MegaMoE compute behind an explicit group16 capability gate until the SFB/TMEM/MMA scale path is updated and validated.

Tested: PYTHONPYCACHEPREFIX=/private/tmp/deepgemm_pycache python3 -m py_compile deep_gemm/mega/__init__.py tests/test_mega_moe.py tests/generators.py

Tested: git diff --check

Not-tested: CUDA build and SM100/B300 runtime validation are not available locally.

csrc/jit_kernels/impls/smxx_layout.hpp

@@ -228,7 +228,7 @@ static torch::Tensor get_k_grouped_mn_major_tma_aligned_packed_ue8m0_tensor(cons
                                                                             const torch::Tensor& ks_tensor,
                                                                             const std::vector<int>& ks,
                                                                             const int gran_k) {
-    DG_HOST_ASSERT(gran_k == 32 or gran_k == 128);
+    DG_HOST_ASSERT(gran_k == 16 or gran_k == 32 or gran_k == 128);
     const auto [sf_k, mn] = get_shape<2>(sf);
     const auto num_groups = static_cast<int>(ks.size());