The system needs time for an IRQ to happen before we are able to see a
change in the HMQ statistics. Waiting 10ms seems reasonable.

Increase also the timeout on the firmware side to 1 second. With these
tests, we are interested in the functionality and not (at least today)
to its performance.

However this test was working fine in the past. For some reason now
there a component that is slower, or faster, than before.

For sure the Python interpreter is different and it could be that is now
much faster compared to some years ago when probably we tested this with
Python 2.7.
Signed-off-by: Federico Vaga <federico.vaga@cern.ch>