pub fn tokenize(text: &str) -> Vec<String>Expand description
Server-default text pipeline: word tokenizer (split on non-alphanumeric), Unicode lowercase, English stopword removal, English snowball stemming.
Matches WordTokenizer + default TokensProcessor on the Qdrant server —
the same pipeline Qdrant Edge’s EdgeBm25 runs. For other languages and
options see crate::bm25_text::Bm25Pipeline.