[WIP] Add FA3 into CMAKELIST

[hebiao064]

Motivation

Currently cd sgl-kernel && VERBOSE=1 make build will fail

[30/31] /usr/local/cuda/bin/nvcc -forward-unknown-to-host-compiler -DPy_LIMITED_API=0x03090000 -DUSE_C10D_GLOO -DUSE_C10D_NCCL -DUSE_DISTRIBUTED -DUSE_RPC -DUSE_TENSORPIPE -Dcommon_ops_EXPORTS -I/home/jobuser/sglang/sgl-kernel/include -I/home/jobuser/sglang/sgl-kernel/csrc -I/home/jobuser/sglang/sgl-kernel/build/_deps/repo-cutlass-src/include -I/home/jobuser/sglang/sgl-kernel/build/_deps/repo-cutlass-src/tools/util/include -I/home/jobuser/sglang/sgl-kernel/build/_deps/repo-flashinfer-src/include -I/home/jobuser/sglang/sgl-kernel/build/_deps/repo-flashinfer-src/csrc -I/home/jobuser/sglang/sgl-kernel/build/_deps/repo-flash-attention-src/hopper -isystem /home/jobuser/.cache/uv/builds-v0/.tmpOxvmYE/lib/python3.10/site-packages/torch/include -isystem /home/jobuser/.cache/uv/builds-v0/.tmpOxvmYE/lib/python3.10/site-packages/torch/include/torch/csrc/api/include -isystem /export/apps/python/3.10/include/python3.10 -isystem /usr/local/cuda/targets/x86_64-linux/include -DONNX_NAMESPACE=onnx_c2 -gencode arch=compute_90,code=sm_90 -Xcudafe --diag_suppress=cc_clobber_ignored,--diag_suppress=field_without_dll_interface,--diag_suppress=base_class_has_different_dll_interface,--diag_suppress=dll_interface_conflict_none_assumed,--diag_suppress=dll_interface_conflict_dllexport_assumed,--diag_suppress=bad_friend_decl --expt-relaxed-constexpr --expt-extended-lambda -O3 -DNDEBUG -std=c++17 -Xcompiler=-fPIC -DNDEBUG -DOPERATOR_NAMESPACE=sgl-kernel -O3 -Xcompiler -fPIC -gencode=arch=compute_75,code=sm_75 -gencode=arch=compute_80,code=sm_80 -gencode=arch=compute_89,code=sm_89 -gencode=arch=compute_90,code=sm_90 -std=c++17 -DFLASHINFER_ENABLE_F16 -DCUTLASS_ENABLE_TENSOR_CORE_MMA=1 -DCUTLASS_VERSIONS_GENERATED -DCUTE_USE_PACKED_TUPLE=1 -DCUTLASS_TEST_LEVEL=0 -DCUTLASS_TEST_ENABLE_CACHED_RESULTS=1 -DCUTLASS_DEBUG_TRACE_LEVEL=0 --expt-relaxed-constexpr -Xcompiler=-Wconversion -Xcompiler=-fno-strict-aliasing -use_fast_math -gencode=arch=compute_90a,code=sm_90a -DFLASHINFER_ENABLE_BF16 -DFLASHINFER_ENABLE_FP8 -DFLASHINFER_ENABLE_FP8_E4M3 -DFLASHINFER_ENABLE_FP8_E5M2 -D_GLIBCXX_USE_CXX11_ABI=0 -MD -MT CMakeFiles/common_ops.dir/csrc/gemm/fp8_gemm_kernel.cu.o -MF CMakeFiles/common_ops.dir/csrc/gemm/fp8_gemm_kernel.cu.o.d -x cu -c /home/jobuser/sglang/sgl-kernel/csrc/gemm/fp8_gemm_kernel.cu -o CMakeFiles/common_ops.dir/csrc/gemm/fp8_gemm_kernel.cu.o
ninja: build stopped: subcommand failed.


*** CMake build failed
  x Failed to build `/home/jobuser/sglang/sgl-kernel`
  |-> The build backend returned an error
  `-> Call to `scikit_build_core.build.build_wheel` failed (exit status: 1)
      hint: This usually indicates a problem with the package or the build
      environment.
make: *** [Makefile:29: build] Error 2

Modifications

Checklist

• Format your code according to the Code Formatting with Pre-Commit.
• Add unit tests as outlined in the Running Unit Tests.
• Update documentation / docstrings / example tutorials as needed, according to Writing Documentation.
• Provide throughput / latency benchmark results and accuracy evaluation results as needed, according to Benchmark and Profiling and Accuracy Results.
• For reviewers: If you haven't made any contributions to this PR and are only assisting with merging the main branch, please remove yourself as a co-author when merging the PR.
• Please feel free to join our Slack channel at https://slack.sglang.ai to discuss your PR.

[hebiao064]

Duplicated with #4902, close it for now