llm, add model section back to qwen settings in case that's why it's freaking out.

tasks/workstation/linux/software/llm.yml

@@ -901,6 +901,10 @@
             "env": {
               "OLLAMA_API_KEY": "ollama"
             },
+            "model": {
+              "name": "qwen2.5-coder:3b",
+              "baseUrl": "http://127.0.0.1:11434/v1"
+            },
             "modelProviders": {
               "openai": [
                 {% for llm in llm_installs %}