fix: correct vLLM service port to 8000

- Updated qwen-2.5-7b port from 8001 to 8000 in models.yaml
- Matches actual vLLM server default port configuration
- Tested and verified: orchestrator successfully loaded model and generated response

model-orchestrator/models.yaml

@@ -7,7 +7,7 @@ models:
     type: text
     framework: vllm
     service_script: models/vllm/server.py
-    port: 8001
+    port: 8000
     vram_gb: 14
     startup_time_seconds: 120
     endpoint: /v1/chat/completions