add mla_preprocess kernel (#3226)

### What this PR does / why we need it?

- Adds the `mla_preprocess` custom kernel to provide an optimized
pre-processing operator for Multi-head Latent Attention (MLA) on Ascend
NPUs.
- Wires the new kernel into the C++ extension pipeline so vLLM can
invoke it directly, cutting Python-side tensor shuffling and memory
copies that previously bottlenecked MLA compilation paths.

### Does this PR introduce any user-facing change?

- No. The change only introduces a low-level kernel; public APIs and
inference behavior remain unchanged.

### How was this patch tested?

- Dedicated Ascend kernels are not covered by our CI yet, so no extra
automated tests were added. Future MLA-focused regression runs will
cover this path.

- vLLM version: v0.11.0

Signed-off-by: Chen Chen <0109chenchen@gmail.com>

CMakeLists.txt

@@ -44,11 +44,13 @@ else()
 endif()
 
 include(${ASCENDC_CMAKE_DIR}/ascendc.cmake)
+
 file(GLOB KERNEL_FILES
 ${CMAKE_CURRENT_SOURCE_DIR}/csrc/kernels/*.cpp)
 
 ascendc_library(vllm_ascend_kernels SHARED
     ${KERNEL_FILES}
+    ${CMAKE_CURRENT_SOURCE_DIR}/csrc/mla_preprocess/op_kernel/mla_preprocess_kernel.cpp
 )
 
 message("TORCH_NPU_PATH is ${TORCH_NPU_PATH}")
@@ -90,7 +92,11 @@ target_link_libraries(
   libtorch_npu.so
   vllm_ascend_kernels
   ascendcl
+  tiling_api
+  register
   platform
+  ascendalog
+  dl
 )
 
 target_link_options(vllm_ascend_C PRIVATE "-Wl,-rpath,$ORIGIN:$ORIGIN/lib")