Make info level logs more clear and consistent

I think the operator should see, at info level:

see one line when it fails (with the nick/fp of both relays and the reason)
see one line when it doesn't (with the nick/fp of both relays) Currently the failures are logged at several po ints in the code (and several levels) and the messages don't provide the same available information

For instance, every time a relay fails to be measured, all these lines are shown:

2018-12-05 12:35:12,104 WARNING Thread-5 circuitbuilder.py:83 - _build_circuit_impl - Circuit failed to be created: CHANNEL_CLOSED
2018-12-05 12:35:12,107 WARNING Thread-5 circuitbuilder.py:83 - _build_circuit_impl - EXTENDCIRCUIT returned unexpected response code: 551
2018-12-05 12:35:12,108 WARNING Thread-5 circuitbuilder.py:83 - _build_circuit_impl - EXTENDCIRCUIT returned unexpected response code: 551
2018-12-05 12:35:12,109 WARNING Thread-5 scanner.py:203 - measure_relay - Could not build circuit involving UbuntuCore250
2018-12-05 12:36:13,848 WARNING Thread-5 scanner.py:217 - measure_relay - When measuring default 565B07B9 the destination seemed to have stopped )

I think this should be only 1 line a level INFO, and the rest at DEBUG level

For success measurements, it logs

2018-12-05 12:40:13,947 INFO Thread-1 resultdump.py:544 - handle_result - Torrr 96EA4E2C finished measurement with ResultSuccess

Instead of ResultSuccess, should say which two relays were measured.

Edit: format