Switched logging to slf4j + logback.

[eupakhomov]

See links below for reasons:
http://javarevisited.blogspot.de/2013/08/why-use-sl4j-over-log4j-for-logging-in.html
https://logback.qos.ch/reasonsToSwitch.html

[coveralls]

Coverage increased (+0.02%) to 57.273% when pulling f71ba02 on eupakhomov:0.5_switch_logging into 72e5393 on ChargeTimeEU:master.

[TVolden]

Code looks fine. :)

[TVolden]

I checked the IC, it's an unstable integration test (json remote stop)

[eupakhomov]

I haven't yet dig into the issue with IC but at firt glance it seems server doesn't start up by the time client expect to. Maybe solution would be to add something like Thread.sleep() or awaitility await() to make it more stable.

[TVolden]

This is what I tried to do here: https://github.com/ChargeTimeEU/Java-OCA-OCPP/blob/master/ocpp-v1_6-test/src/test/groovy/eu/chargetime/ocpp/test/core/json/JSONRemoteStopTransactionSpec.groovy#L30-L34

I suspect it's related to the order of the tests, the jsonRemoteStopTransaction is the first test to run after a series of SOAP tests. I can make the test pass, just by restarting the job on Travis.
It's not a constant problem, but mostly isolated to the IC server. I have a local jenkins running without the problem.