Add per exon p-value and log2 fold changes to exon Excel output

[tjakobi]

Right now the Excel file could provide more detailed information that is scattered throughout different files. This information should be unified in the produced .XLS file.

Bonus points: can we also incorporate newly detected exons from the reconstruct/FUCHS module?

[tjakobi]

Addressing the bonus point of FUCHS exons: This might be a more complex task since the script directly reads the ballgown output and currently does not add any other annotation to the data. I assume we would have to re-create the ballgown data with the added exons first instead of try to merge information later.