Timeouts in user defined handlers should be treated the same way as other exceptions raised from within the user handler.
This should cause retry.

Currently the request that times out will end up in strange state in RequestQueue and it will block crawler from closing for 300 secs until following triggers:
"The request queue seems to be stuck for 300.0s, resetting internal state."

(Regarding blocking sync code timeouts from user defined handlers: #908)