Now input vector with class is required for unsupervised classifier.
The cases where input vector data were optional for unsupervised
classification are commented.

Update license for added/moved files.

This allow to have only one application to handle bot supervised and
unsupervised classifier.

Classifier category is retrieve with selected classifier and depending
this category some operation are added/removed for the training.

A choice is used to select between contingency table and confusion
matrix.

SampleSelection can now be used without input vector data.

When only image is provided to the application a default vector
corresponding to the image envelope is generated.

ENH: Move the new deburst filter to SARCalibration module and remove the SAR module introduced in this RFC (from RFC review)

Use inheritance with otbTrainImagesBase to create unsupervised
composite application.

Create a common class with template to perform Supervised classification
(same as otbTrainImagesClassifier) or Unsupervised.

Add the ability to specify Supervised or Unsupervised classification

Inherit TrainVectorClassifier from TrainVectorBase and use Non-Virtual
Function Idiom to provide common behavior for Unsupervised and
Supervised classification.

Move commons files for both supervised and unsupervised into
LearningBase Module.
Also update Test and includes.

Change typedef to use otbWrapperTypes ones.
AddProcess for the applications.
Add progress Report for otbGeodesicMorphology* filters
Add documentation limits.
Remove unnecessary include and Update* call.