Methods are tested in specific conditions:
* at first position for countBefore
* at last position for countAfter
* when the affectation doesn't exist in the sequence

A hard-coded sequence should not be given to the constructor, instead it must be stored into a variable.

Variables removed from affect_infos_s. Can't be tested anymore

When the k-mer ends at the position we're considering, it
should be taken into account by default.

On failure firstMax and lastMax now return -1

By default the allowed overlap is now 0 but tests were not designed
with this overlap (it was somehow s-1). So we set the default overlap
to this value, so that the tests don't have to change.

We'll test the overlap in other tests.

	* Refactored main vidjil.cpp (core/windows.cpp, core/windowExtractor.cpp)
	* Removed unused html output
	* Better json output (core/json.cpp)
	* Updated main stdout, with representative sequence for each clone
	* Updated parameters for FineSegmenter (delta_max) and dynprog (substition cost)
	* Bugs closed

	* First public release