Feature: Adding Metric Aggregation in the form of a StatisticSet

[lksfrnz]

Introduces a new metric type called StatisticSet which will aggregate a set of metrics into the following properties: maximum, minimum, count, and sum. Users can create and log their own StatisticSet metric with the StatisticSetBuilder class or specify a metric's method when putting it into the metricsLogger and the metricsLogger will handle all aggregation.

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

[giancarlokc]

Why are we returning null here?

[giancarlokc]

Does it make sense to return the own object here? Doesn't seem to actual implement what the method name suggests nor does it use the method's arguments.

[lksfrnz]

Using a PriorityQueue here instead of just cycling through a queue is a slightly slower implementation but ensures that the minimum amount of metrics are flushed which the previous solution didn't.

For example:
let MAX_METRICS_PER_EVENT = 2
let MAX_DATAPOINTS_PER_METRIC = 1

lets say we have a queue of metrics represented as how data points are in each metric
{"M1" : 1, "M2" : 1, "M3" : 2}

The previous implementation would result in 3 flushes as follows
Flushed : ["M1", "M2"] Remaining : {"M3" : 2}
Flushed : ["M3"] Remaining : {"M3" : 1}
Flushed : ["M3"] Remaining : {}

The New implementation will result in 2 flushes as follows:
Flushed : ["M1", "M3"] Remaining : {"M2" : 1, "M3" : 1}
Flushed : ["M2", "M3"] Remaining : {}