[Feature Request] Improve indexing performance by using a single Lucene field

[bugmakerrrrrr]

Is your feature request related to a problem? Please describe

Currently, OS usually create multiply Lucene fields for a single OS field during indexing, for example, when parsing number field, OS creates three Lucene fields.

When Lucene processes one document, it needs two passes for each fields. The first pass is to verify that doc schema matches the index schema, and the second pass is to index each field. Obviously the more fields there are, the more passes need to be processed.

Describe the solution you'd like

For each OS field, only create a single Lucene field to reduce the number of passes to be processed. Take a Long field as a example, we can create a Lucene field as follows:

final FieldType type = new FieldType();
type.setDimensions(1, Long.BYTES);
type.setDocValuesType(DocValuesType.SORTED_NUMERIC);
type.setStored(true);

new CustomField(name, value, type)

Related component

Indexing:Performance

Describe alternatives you've considered

No response

Additional context

No response