Add LD_LIBRARY_PATH to GitHub Actions workflows for CUDA library discovery

Set LD_LIBRARY_PATH in all workflow steps to ensure CUDA 11.4 libraries
are found during both compile time and runtime. This fixes the issue where
the GGML CUDA backend (libggml-cuda.so) fails to load when running
'ollama serve'.

Library paths added:
- /usr/local/cuda-11.4/lib64
- /usr/local/cuda-11.4/targets/x86_64-linux/lib
- /usr/lib64
- /usr/local/lib64

Updated workflows:
- tesla-k80-ci.yml: CMake configure, C++/CUDA build, Go build, binary verify
- tesla-k80-single-gpu-tests.yml: All test execution steps
- tesla-k80-multi-gpu-tests.yml: All test execution steps

.github/workflows/tesla-k80-ci.yml

@@ -28,20 +28,27 @@ jobs:
           CC=/usr/local/bin/gcc CXX=/usr/local/bin/g++ cmake -B build
         env:
           CMAKE_BUILD_TYPE: Release
+          LD_LIBRARY_PATH: /usr/local/cuda-11.4/lib64:/usr/local/cuda-11.4/targets/x86_64-linux/lib:/usr/lib64:/usr/local/lib64
 
       - name: Build C++/CUDA components
         run: |
           CC=/usr/local/bin/gcc CXX=/usr/local/bin/g++ cmake --build build -j$(nproc)
         timeout-minutes: 30
+        env:
+          LD_LIBRARY_PATH: /usr/local/cuda-11.4/lib64:/usr/local/cuda-11.4/targets/x86_64-linux/lib:/usr/lib64:/usr/local/lib64
 
       - name: Build Go binary
         run: |
           go build -v -o ollama .
+        env:
+          LD_LIBRARY_PATH: /usr/local/cuda-11.4/lib64:/usr/local/cuda-11.4/targets/x86_64-linux/lib:/usr/lib64:/usr/local/lib64
 
       - name: Verify binary was created
         run: |
           ls -lh ollama
           ./ollama --version
+        env:
+          LD_LIBRARY_PATH: /usr/local/cuda-11.4/lib64:/usr/local/cuda-11.4/targets/x86_64-linux/lib:/usr/lib64:/usr/local/lib64
 
       - name: Upload ollama binary as artifact
         uses: actions/upload-artifact@v4

.github/workflows/tesla-k80-multi-gpu-tests.yml

@@ -30,6 +30,8 @@ jobs:
           chmod +x ollama
           ls -lh ollama
           ./ollama --version
+        env:
+          LD_LIBRARY_PATH: /usr/local/cuda-11.4/lib64:/usr/local/cuda-11.4/targets/x86_64-linux/lib:/usr/lib64:/usr/local/lib64
 
       - name: Verify multi-GPU setup
         run: |
@@ -49,15 +51,21 @@ jobs:
           go build -o ../../test-runner .
           cd ../..
           ls -lh test-runner
+        env:
+          LD_LIBRARY_PATH: /usr/local/cuda-11.4/lib64:/usr/local/cuda-11.4/targets/x86_64-linux/lib:/usr/lib64:/usr/local/lib64
 
       - name: Validate multi-GPU test configuration
         run: |
           ./test-runner validate --config test/config/models.yaml --profile multi-gpu
+        env:
+          LD_LIBRARY_PATH: /usr/local/cuda-11.4/lib64:/usr/local/cuda-11.4/targets/x86_64-linux/lib:/usr/lib64:/usr/local/lib64
 
       - name: Run multi-GPU tests
         run: |
           ./test-runner run --profile multi-gpu --config test/config/models.yaml --output test-report-multi-gpu --verbose
         timeout-minutes: 60
+        env:
+          LD_LIBRARY_PATH: /usr/local/cuda-11.4/lib64:/usr/local/cuda-11.4/targets/x86_64-linux/lib:/usr/lib64:/usr/local/lib64
 
       - name: Check multi-GPU test results
         run: |

.github/workflows/tesla-k80-single-gpu-tests.yml

@@ -27,6 +27,8 @@ jobs:
           chmod +x ollama
           ls -lh ollama
           ./ollama --version
+        env:
+          LD_LIBRARY_PATH: /usr/local/cuda-11.4/lib64:/usr/local/cuda-11.4/targets/x86_64-linux/lib:/usr/lib64:/usr/local/lib64
 
       - name: Build test-runner
         run: |
@@ -36,15 +38,21 @@ jobs:
           go build -o ../../test-runner .
           cd ../..
           ls -lh test-runner
+        env:
+          LD_LIBRARY_PATH: /usr/local/cuda-11.4/lib64:/usr/local/cuda-11.4/targets/x86_64-linux/lib:/usr/lib64:/usr/local/lib64
 
       - name: Validate test configuration
         run: |
           ./test-runner validate --config test/config/quick.yaml
+        env:
+          LD_LIBRARY_PATH: /usr/local/cuda-11.4/lib64:/usr/local/cuda-11.4/targets/x86_64-linux/lib:/usr/lib64:/usr/local/lib64
 
       - name: Run quick tests
         run: |
           ./test-runner run --profile quick --config test/config/quick.yaml --output test-report-quick --verbose
         timeout-minutes: 10
+        env:
+          LD_LIBRARY_PATH: /usr/local/cuda-11.4/lib64:/usr/local/cuda-11.4/targets/x86_64-linux/lib:/usr/lib64:/usr/local/lib64
 
       - name: Check quick test results
         run: |
@@ -71,6 +79,8 @@ jobs:
         run: |
           ./test-runner run --profile full --config test/config/models.yaml --output test-report-full --verbose
         timeout-minutes: 45
+        env:
+          LD_LIBRARY_PATH: /usr/local/cuda-11.4/lib64:/usr/local/cuda-11.4/targets/x86_64-linux/lib:/usr/lib64:/usr/local/lib64
 
       - name: Check full test results
         if: success()