HNSW Support for Vector Search

[loretoparisi]

Is your feature request related to a problem? Please describe.
Indexing vectors of embeddings along with the document. Optionally supporto multi-vector per document and retrieval.

Describe the solution you'd like
Add HNSW as vector similarity search

Describe alternatives you've considered

• OpenSearch >= 8
• Vespa
• Vector Stores (Pinecone, ChromaDB, etc.)

Additional context
Semantic and Similarity Search integration to keyword based search.

[sanikolaev]

The following SQL syntax is proposed for the new field:

<field name> 
  float_vector 
    [knn_type='hnsw'
      knn_dims='int'
      knn_similarity={l2|ip|cosine}
      [hnsw_m='int']
      [hnsw_ef_construction='int']
    ]

• knn_type is not mandatory. If no knn* is specified, the field remains just an array of floats
• knn_type gets turns on automatically if knn_similarity or knn_dims is specified. The default is hnsw.
• knn_dims and knn_similarity are required if knn_type='hnsw'
• hnsw_m and hnsw_ef_construction are optional

Examples:

• create table t(a float_vector) - just an array of floats
• create table t(a float_vector knn_dims='128' knn_similarity='l2') - simplest syntax to enable knn
• create table t(a float_vector knn_type='hnsw' knn_dims='128' knn_similarity='l2') - alternative syntax mostly for the future when knn_type can be e.g. annoy
• create table t(a float_vector knn_type='hnsw' knn_dims='16' knn_similarity='ip' hnsw_m='16') - fine-tuning
• create table t(a float_vector knn_type='hnsw' knn_dims='16' knn_similarity='ip' hnsw_m='20' hnsw_ef_construction='90') - more fine-tuning

@glookka pls review and let me know if it looks good or if I'm missing something and there are better options.

[glookka]

knn_similarity={l2|ip|cosine} option is specific to HNSW. E.g. annoy has "angular", "euclidean", "manhattan", "hamming", or "dot". So it probably makes sense to name the option hnsw_similarity.

[sanikolaev]

Closing as done - https://manual.manticoresearch.com/dev/Searching/KNN#K-nearest-neighbor-search.