text_match score value is not affected by query_by_weights values

Description

According to the documentation query_by_weights parameter is a way to give a relative importance to different fields

The relative weight to give each query_by field when ranking results. Values can be between 0 and 127. This can be used to boost fields in priority, when looking for matches.

In practice, higher query_by_weights does not result in higher _text_match.

Prior to v.26.0 ranking was also random, query_by_weights was not considered at all; in 26.0 on anecdotal cases it seems, that ranking expecation is met (needs to be confirmed on larger sample set, since _text_match value is still the same,independent of which fields were hit and which query_by_weights values they had.

Steps to reproduce

Request:

"request": {
"q": "san diego",
"max_hits": 20,
"per_page": 20,
"prioritize_exact_match": "false",
"prioritize_token_position": "true",
"min_len_1typo": 3,
"min_len_2typo": 7,
"use_cache": "false",
"enable_typos_for_numerical_tokens": "false",
"enable_lazy_filter": "true",
"prioritize_num_matching_fields": "false",
"max_candidates": "40",
"sort_by": "_text_match:desc, location(35.10193405724608,-119.88281250000001,precision: 5 km):asc, popularity_bucket:desc",
"drop_tokens_threshold": 5,
"query_by": "geo_preferred_name,admin_name",
"query_by_weights": "5,1",
"num_typos": "2,1",
"prefix": "true,true",
"collection": "geo_US"
}

Document #1:

"document": {
"EntityType": "Municipality",
"admin_name": ["San Diego","CA","California"],
"admin_name_EN": "San Diego,San Diego,CA",
"geo_preferred_name": ["San Diego"],
"id": "f-BFQbwnssVlw3gjoodFKQ",
"location": [32.71575927734375,-117.163818359375],
"popularity_bucket": 3,
}

Document #2:

"document": {
"EntityType": "Municipality",
"admin_name": ["San Diego","CA","Oceanside","California"
],
"admin_name_EN": "Oceanside,San Diego,CA",
"geo_preferred_name": ["Oceanside"],
"geohash_3": "8329a4fffffffff",
"id": "WuqSAEhdDOi30dW5tkOQfg",
"location": [33.19590759277344,-117.3795166015625],
"popularity_bucket": 3,
}

Expected Behavior

text_match of document #1 is higher than of document #2
document #1 is ranked higher than document #2

Actual Behavior

text_match of document #2 and document #1 is the same.
document #1 is ranked higher than document #2 , which is an improvement, comparing to v. 0.25.x

Document 1 text match:
"text_match": 1157451471441621000,
"text_match_info": {
"best_field_score": "2211897868542",
"best_field_weight": 5,
"fields_matched": 0,
"num_tokens_dropped": 0,
"score": "1157451471441621032",
"tokens_matched": 2,
"typo_prefix_score": 0
},

Document 2 text match:
"text_match": 1157451471441621000,
"text_match_info": {
"best_field_score": "2211897868542",
"best_field_weight": 1,
"fields_matched": 0,
"num_tokens_dropped": 0,
"score": "1157451471441621000",
"tokens_matched": 2,
"typo_prefix_score": 0
},

Possibly text_match.score is used in sorting (and its calculation seems to be more consistent, comparing to v.0.25.x), still text_match value output is confusing,

Metadata

Typesense Version: 26.0

OS: Linux

No source file or test is named. Reproduce the Typesense 26.0 request with the two supplied documents, then inspect how query_by_weights contributes to text_match and text_match_info; done means the reported score reflects the differing field weights while document 1 remains ranked higher.