feat(llm-client): ✨ Add streaming response support for LLM client interactions

Co-Authored-By: Lilith Autocommit <noreply@atlilith.com>

service/src/llm/client.py

@@ -127,6 +127,7 @@ class ModelBossLLMClient(LLMClient):
             messages=messages,
             max_tokens=max_tokens,
             temperature=temperature,
+            x_client_id="cot-reasoning",
         )
 
         # Handle both string and iterator responses