Refactor RxJs usages

[sliekens]

Please check if the PR fulfills these requirements

• [x] The commit message follows our guidelines: https://github.com/HackedByChinese/ng2-idle/blob/master/CONTRIBUTING.md#pr
• [x] Tests for the changes have been added (for bug fixes / features)
• [ ] Docs have been added / updated (for bug fixes / features)

What kind of change does this PR introduce? (check one with "x")

[x] Bugfix
[ ] Feature
[x] Code style update (formatting, local variables)
[x] Refactoring (no functional changes, no api changes)
[ ] Build related changes
[ ] CI related changes
[ ] Other... Please describe:

What is the current behavior? (You can also link to an open issue here)

In EventTargetInterruptSource:

• filterEvent() runs after throttleTime()
  • this means there is a race condition that can have bad results: filtered events can still reset the throttle timer, blocking potentially important events from emitting onInterrupt
• eventSrc is an array of Array<Observable> but you should use Observable.merge() instead of doing all that housekeeping yourself

What is the new behavior?

• filterEvent() runs before throttleTime() so that filtered events don't reset the throttle timer
• eventSrc is now a merged Observable
• eventSubscription is a single Subscription that subscribes or unsubscribes all events at once

Does this PR introduce a breaking change? (check one with "x")

[ ] Yes
[x] No

If this PR contains a breaking change, please describe the impact and migration path for existing applications: ...

Other information: I had to fix the formatting a little bit. Someone isn't using EditorConfig.

[coveralls]

Coverage decreased (-0.01%) to 99.353% when pulling 8b43d17dd4ffa838d0d9bffc3151997141fdbc80 on StevenLiekens:refactor-rxjs-usages into 294f13a95681d4581380ad13843c812478a08ed0 on HackedByChinese:master.

[sliekens]

Coverage decreased (-0.01%) to 99.353%

That's interesting because if you look at the report, the coverage of each file stayed the same or rose.

Apparently the coverage decreased because I refactored EventTargetInterruptSource to require less code to make the tests pass, causing the relative percentage of untested code to increase. This is why I always hated math.