fix: log Ollama disconnections with specific error handling (#92)

- BaseAgent.run(): catch httpx.ConnectError/ReadError/ConnectionError,
  log 'Ollama disconnected: <error>' at ERROR level, then re-raise
- session.py: distinguish Ollama disconnects from other errors in
  chat(), chat_with_tools(), continue_chat() — return specific message
  'Ollama appears to be disconnected' instead of generic error
- 11 new tests covering all disconnect paths

src/timmy/session.py

@@ -11,6 +11,8 @@ let Agno's session_id mechanism handle conversation continuity.
 import logging
 import re
 
+import httpx
+
 logger = logging.getLogger(__name__)
 
 # Default session ID for the dashboard (stable across requests)
@@ -83,6 +85,9 @@ async def chat(message: str, session_id: str | None = None) -> str:
     try:
         run = await agent.arun(message, stream=False, session_id=sid)
         response_text = run.content if hasattr(run, "content") else str(run)
+    except (httpx.ConnectError, httpx.ReadError, ConnectionError) as exc:
+        logger.error("Ollama disconnected: %s", exc)
+        return "Ollama appears to be disconnected. Check that ollama serve is running."
     except Exception as exc:
         logger.error("Session: agent.arun() failed: %s", exc)
         return "I'm having trouble reaching my language model right now. Please try again shortly."
@@ -111,6 +116,11 @@ async def chat_with_tools(message: str, session_id: str | None = None):
 
     try:
         return await agent.arun(message, stream=False, session_id=sid)
+    except (httpx.ConnectError, httpx.ReadError, ConnectionError) as exc:
+        logger.error("Ollama disconnected: %s", exc)
+        return _ErrorRunOutput(
+            "Ollama appears to be disconnected. Check that ollama serve is running."
+        )
     except Exception as exc:
         logger.error("Session: agent.arun() failed: %s", exc)
         # Return a duck-typed object that callers can handle uniformly
@@ -133,6 +143,11 @@ async def continue_chat(run_output, session_id: str | None = None):
 
     try:
         return await agent.acontinue_run(run_response=run_output, stream=False, session_id=sid)
+    except (httpx.ConnectError, httpx.ReadError, ConnectionError) as exc:
+        logger.error("Ollama disconnected: %s", exc)
+        return _ErrorRunOutput(
+            "Ollama appears to be disconnected. Check that ollama serve is running."
+        )
     except Exception as exc:
         logger.error("Session: agent.acontinue_run() failed: %s", exc)
         return _ErrorRunOutput(f"Error continuing run: {exc}")