https://github.com/intel/ScalableVectorSearch/pull/97 will add a first version of `SQDataset`. Some improvements will be necessary. * Performance benchmarking on large datasets should be done, bottlenecks identified and fixed. * Documentation needs to be updated. * Python bindings.
