Clean up server_args.py to have a dedicated function for model specific adjustments (#8983)

.github/workflows/vllm-dependency-test.yml

@@ -30,11 +30,12 @@ jobs:
       - name: Install dependencies
         run: |
           bash scripts/ci_install_dependency.sh
-          pip install "vllm==0.10.0"
+          pip install "vllm==0.9.0"
           pip install "bitsandbytes>=0.44.0"
-          pip3 install torch==2.8.0 torchvision torchaudio --index-url https://download.pytorch.org/whl/test/cu126
+          pip install torch==2.8.0 torchvision torchaudio --index-url https://download.pytorch.org/whl/test/cu126
+          pip install "openai==1.99.1"
 
-      - name: Run VLLM dependency tests
+      - name: Run vLLM dependency tests
         timeout-minutes: 60
         run: |
           cd test/srt