Release of 2.1.1 (bug fix release)

A last minute bug sneaked into version 2.1.0 (just released yesterday). It was triggered when calling SNVs and indels simultaneously with lofreq call (not somatic) and would have resulted in a segfault in most cases (at the automatically run filtering stage).

The actual reason was broken indexing during the FDR correction for strand-bias filtering when indels and SNVs where present in the same vcf file. The somatic pipeline was not affected by this because it keeps separate files for SNVs and indels.

Andreas