kundajelab/tfmodisco: Reverting to previous defaults for seqlet identification, some minor fixes

Corresponds to PR https://github.com/kundajelab/tfmodisco/pull/82 Default settings for variable-length seqlet identification seem not-great, so reverting to the previous default of fixed-length seqlet identification (which has been battle-tested more) Putting in the fix caught by #81 Another minor fix for cases where there are lots of ties in the scores during seqlet identification (which could occur with the variable-length seqlet identification