Deduct overall smeshing statuses from events

[brusherru]

In case we got some problems the application should "scream" about the problem and display some big error message.

For example, if Smesher service cannot compute the PoS proof in time, it should be noticed by the User to be able to fix the problem before the next round starts.

To do so we need to determine which events/statuses might be the indicator that something goes wrong and what to display.

It is connected to the issue of displaying identity states, so #4 is prerequisite.