[Bugfix] fix custom op GmmSwigluQuantWeightNzTensorList (#4593)

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

1. Fixes the environment path used to locate custom op shared libraries.
2. Uses empty tensor initialization for op outputs instead of
zero-initialization for better efficiency.



- vLLM version: v0.11.2
- vLLM main: https://github.com/vllm-project/vllm/commit/v0.11.2

---------

Signed-off-by: QianChenxi <chenxi.qian.cq@outlook.com>

csrc/CMakeLists.txt

@@ -15,7 +15,7 @@ option(BUILD_OPEN_PROJECT         "Build open ascend ops project."  ON)
 option(ENABLE_CCACHE              "Enable ccache capability"        ON)
 set(ASCEND_COMPUTE_UNIT           "ascend910b"                    CACHE   STRING   "soc that need to be compiled")
 set(ASCEND_OP_NAME                "ALL"                           CACHE   STRING   "operators that need to be compiled")
-set(VENDOR_NAME                   "customize"                     CACHE   STRING   "vendor name")
+set(VENDOR_NAME                   "vllm-ascend"                   CACHE   STRING   "vendor name")
 
 include(cmake/config.cmake)
 include(cmake/func.cmake)