Compute Array Statistics Enhancemets

[imikejackson]

• Add option to also output an array that has the Feature id in it.

• Do not fill in from zero to min feature id. That could go bad if someone has feature Ids that start at 1 billion.

• The created group needs to be an Attribute Matrix and not a Group. This will break a unit test because computing the histogram for an array does not produce a matching tuple and component shape.

• Option to set the "Feature Id" range.
  This would allow a user to set a range of feature ids to calculate statistics for.
  Allows the user to "pad out the feature ids" to a specific range

• Option to Ignore Feature Id Zero.

• The creation of the attribute Matrix will pad from zero to the minimum feature id that is encountered in the Feature ID Array. This could have interesting side effects of blowing up the Attribute Matrix if the minimum Feature Id is something like 2 billion