Fix resume runtime after a pause

[rbren]

End-user friendly description of the problem this fixes or functionality that this introduces

• Include this change in the Release Notes. If checked, you must provide an end-user friendly description for your change below
  no changelog

---

Give a summary of what the PR does, explaining any non-trivial design decisions

If you hang out in your session too long, the backend will pause your runtime. If you then send a new message to the agent, you get a 503 error.

This tries to resume the runtime whenever we get a 503

This might not be the best design--probably better to put the agent back into INITIALIZING state so that the user can see it warming up. But this works a little better for now

---

Link of any specific issues this addresses

---

To run this PR locally, use the following command:

docker run -it --rm   -p 3000:3000   -v /var/run/docker.sock:/var/run/docker.sock   --add-host host.docker.internal:host-gateway   -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:e577e02-nikolaik   --name openhands-app-e577e02   docker.all-hands.dev/all-hands-ai/openhands:e577e02