fix: handle HTTP redirects in event stream and improve JWT expiration handling

backend/app/main.py

@@ -47,8 +47,25 @@ async def event_stream(request: ChatRequest) -> AsyncGenerator[str, None]:
         "Accept": "text/event-stream",
     }
 
-    async with httpx.AsyncClient(timeout=None, follow_redirects=True) as client:
+    async with httpx.AsyncClient(timeout=None, follow_redirects=False) as client:
         async with client.stream("POST", OLLAMA_API_URL, json=payload, headers=headers) as response:
+            if response.status_code in (301, 302, 307, 308) and response.headers.get("location"):
+                location = response.headers["location"]
+                await response.aclose()
+                logger.debug("Following Ollama redirect to %s", location)
+                async with client.stream("POST", location, json=payload, headers=headers, follow_redirects=False) as response:
+                    if response.status_code >= 400:
+                        text = await response.aread()
+                        raise HTTPException(status_code=502, detail=f"Ollama Cloud error: {text.decode('utf-8', 'ignore')}" )
+
+                    async for chunk in response.aiter_text():
+                        if not chunk:
+                            continue
+                        for line in chunk.splitlines():
+                            if line.strip():
+                                yield f"data: {line}\n\n"
+                return
+
             if response.status_code >= 400:
                 text = await response.aread()
                 raise HTTPException(status_code=502, detail=f"Ollama Cloud error: {text.decode('utf-8', 'ignore')}" )