No connection attempts after the paired device's mDNS entry expired

[randomdudebunchofnumbers]

Hello,

this pull request tries to solve issue #73. I'd like to ask you to have a look and tell me what you think. Thank you very much.

[coveralls]

coverage: 94.82% (+0.1%) from 94.71% — randomdudebunchofnumbers:bug/attempt_reconnect_when_device_reappears into enbility:dev

[randomdudebunchofnumbers]

To be fair, the analysis of the current implementation revealed some issues with stale timers and memory leaks. Nothing critical, but worth checking. I've now updated my pull request with some fixes and tests and would really appreciate it if you could have a look. Thank you.

[sthelen-enqs]

We actually discovered and looked into this issue independently early last week, I'm not sure the connectionAttemptGeneration is necessary.
We created a branch that should fix the same issue. We bypass the race condition you mention in your PR by only resetting isConnectionAttemptRunning state if we manage to cancel the Timer before it fires.
Could you evaluate if this bypasses the necessity for the added generations?
I'd like to avoid added complexity if possible, but don't mind either solution as long as they fix the issue (well).