feat(ci): add support for running vision inference tests

.github/actions/setup-vision-ollama/action.yml

@@ -0,0 +1,11 @@
+name: Setup Ollama for Vision Tests
+description: Start Ollama with vision model
+runs:
+  using: "composite"
+  steps:
+    - name: Start Ollama
+      shell: bash
+      run: |
+        docker run -d --name ollama -p 11434:11434 docker.io/llamastack/ollama-with-vision-model
+        echo "Verifying Ollama status..."
+        timeout 30 bash -c 'while ! curl -s -L http://127.0.0.1:11434; do sleep 1 && echo "."; done'