fix(workflow): remove unnecessary LLAMA_ARGS from windows ggml-cuda build step

Author: Gioee

.github/workflows/main.yml

@@ -226,7 +226,7 @@ jobs:
         shell: cmd
         run: |
           call "C:\Program Files\Microsoft Visual Studio\2022\Enterprise\VC\Auxiliary\Build\vcvarsall.bat" x64
-          make build/llama.cpp.stamp ${{ matrix.make && matrix.make || ''}} LLAMA_ARGS="--target ggml-cuda"
+          make build/llama.cpp.stamp ${{ matrix.make && matrix.make || ''}}
 
       - name: windows install hip toolkit
         if: matrix.name == 'windows-hip'