Creation of classes for user parameters (basic and advanced)
Modification of existant classes (Local variables removed to refer to global variables)
Regroupement of classes to avoid code repetition (LocalExtremaDetection)
Removal of unnecessary classes (stored in snippets)

* All the objects `Img<T>` used as parameters have been replaced by `RandomAccessibleInterval<T>` or `IterableInterval<T>` objects.
* A parameter `ImgFactory<T>` has been added when the creation of an `Img<T>` as an output was necessary.
* The class `Histogram<T>` from package `net.imglib2.algorithm.stats` has been copied (class HistogramZ) and the constructor parameter `Img<T> `has been change into `IterableInterval<T>`.

Addition of Input dimensions as instance variables (instead of static variable in SurfacesExtraction class)

Bug fixed : the number of point is based on the actual content of the 2D array and not on the number of Coordinates objects created.

Refactor :  Replacement of X, Y and Z by width, height and depth when referring to the size of the image

 remove of starting OSE size as a static variable and replace by a new class OSEStartingSize

- Code cleanup

For clarification and modularity purposes the method uses for maximums selection are regrouped in one class which produces a 2D pixels array. This array is afterwards used to reconstructed the OS then the surfaces.

The amplitude of maximums are used to classified the pixels. This class is used in complement with LocalOtsu class.