Worker manager refactor

[benblack86]

Main aim was just to improve the logging coming from the worker manager by adding the io system name to each log statement. Whilst I was there I also:

• Increased maxTries for the waitForStatus test method as I was getting random failures on my slow computer.
• Renamed WorkerAgent to the more accurate name of WorkerRefs.
• Moved WorkerFactory out of WorkerManager file.
• Generally refactored the code.

No logic changes! Fixes #619

@aliyakamercan @jlbelmonte @amotamed @dxuhuang @DanSimon

[codecov-io]

Codecov Report

Merging #629 into develop will decrease coverage by 0.01%. The diff coverage is 82.92%.

@@             Coverage Diff             @@
##           develop     #629      +/-   ##
===========================================
- Coverage    84.76%   84.75%   -0.02%     
===========================================
  Files           98       99       +1     
  Lines         4432     4435       +3     
  Branches       366      354      -12     
===========================================
+ Hits          3757     3759       +2     
- Misses         675      676       +1

Impacted Files	Coverage Δ
...lossus/src/main/scala/colossus/core/IOSystem.scala	93.75% <ø> (ø)	:arrow_up:
...s/src/main/scala/colossus/core/WorkerFactory.scala	100% <100%> (ø)
...s/src/main/scala/colossus/core/WorkerManager.scala	86.45% <80.55%> (-1.3%)	:arrow_down:

---

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update e1bb232...ad3cc51. Read the comment docs.

[benblack86]

@amotamed I believe both WorkerFactory and DefaultWorkerFactory should only be used internally, therefore kept private.