fixed unit and non regression test; changed non regression test for the -q option; fully commented the handling of undef files

 bugfix when PE treated as separated, undef filter considered them as single. bugfix when minimum quality score for nucleotides is 0, nucleotides with score 0 were considered as errors. -q option now takes  argumens (see News file and doc)

bugfix on OS X: output file was not overwritten when it already existed. Content was added to it. This is now fixed.

did new implementation for using only the max(k-mer value, reverse complement value) in the CMS; updates unit tests accordingly.

Added 4rth part (component that writes filtered fastq. Added tests for it. Had to refactor a little bit the FqBackends for it.

started writing main program to do some perfrmance testing. Had to refactor alittle bit fqreader and ead_utils components. Added tests accordingly

 started to develop part 2: retrieveing reads and transforming them into a serie of number. To do that, had to refactor a little bit what I had done for part1 : fqreader and backends. Also added some tests for part2 but they are in unit_test_fqreader.cpp for the moment.

 started to develop part 2: retrieveing reads and transforming them into a serie of number. To do that, had to refactor a little bit what I had done for part1 : fqreader and backends. Also added some tests for part2 but they are in unit_test_fqreader.cpp for the moment.


Former-commit-id: d04b888b