Solr doesn't expose Lucene's BooleanSimilarity (ref
LUCENE-5867) so it's not available for use in situations where BM25/TDF-IF are not useful. (Fields using this similarity will likely also set omitNorms and omitTermFreqAndPositions to true.)
Our use case is ngram-driven suggestions, where the frequency of occurrence of a particular sequence of characters is not something users would expect to be taken into account when ordering suggestions.
Here's my PR: https://github.com/apache/lucene-solr/pull/867 (I'm at Activate if anyone would like to talk this through in person.)