Fix modeling/scoring to handle single-sequence input files

[christinehc]

rules.build_model currently assumes multiple sequences per FASTA file, where each FASTA file contains sequences all belonging to the family specified in the filename.

Assessment of model performance is currently conducted via K-fold cross validation, which requires at least 2 samples per class. Thus, single-sequence files will throw a ValueError.

Secondary structure output files will only contain a single sequence, so a rewrite of this rule is required for Snekmer to run on SS output files correctly.

[christinehc]

Bug discovered by @DarrenJSchmidt