perf: compact batched FTS DocID outputs

[zhanhongliu2001]

Summary

• compact batched FTS DocID outputs in the sparse DAAT iterator
• reduce repeated per-row output handling in the FTS retrieval hot path
• submit the exact code state at commit 322594223ea35cba11e898ebfc7b92f5fc6bbcf8 for VLDB benchmark evaluation

Validation

• verified the local branch head exactly matches 322594223ea35cba11e898ebfc7b92f5fc6bbcf8
• git diff --check 3225942^ 3225942 passes
• no additional local generated files are included

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you all sign our Contributor License Agreement before we can accept your contribution.
2 out of 4 committers have signed the CLA.

✅ cao1629
✅ sallyjinx
❌ wangzelin.wzl
❌ zhanhongliu2001

---

wangzelin.wzl seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[LINxiansheng]

Document AI & IK Custom Dictionary Score

Document AI Functions Score
===========================
score: 100.00 / 100
load_file: 50 / 50
ai_split_document: 50 / 50

IK Custom Dictionary Score
==========================
score: 100.00 / 100
ik_custom_dict: 100 / 100

FTS Large Benchmark Score

FTS Large Benchmark Score
=========================
score: 79.85 / 100
mean_improvement: 39.92%
full_score_improvement: 50.00%

build_improvement: 44.11%
  build_ik_all_sec: baseline=35.2836, current=19.799, improvement=43.89%
  build_ik_content_sec: baseline=28.3764, current=15.633, improvement=44.91%
  build_beng_en_sec: baseline=14.7578, current=8.332, improvement=43.54%
tokenize_improvement: -4.67%
  tokenize_ik_avg_ms: baseline=0.76478, current=0.7871, improvement=-2.92%
  tokenize_beng_avg_ms: baseline=0.42262, current=0.4498, improvement=-6.43%
query_improvement: 80.33%
  query_cn_avg_ms: baseline=16.6628, current=3.1315, improvement=81.21%
  query_beng_avg_ms: baseline=24.3042, current=5.9346, improvement=75.58%
  query_mixed_avg_ms: baseline=17.5593, current=4.6886, improvement=73.30%
  query_limit_avg_ms: baseline=16.2334, current=1.4214, improvement=91.24%

FTS Large Benchmark Report

========================================
FTS Large Benchmark Report
========================================
timestamp:              2026-07-18 07:45:46
label:                  vldb-ci-29635160087-1
git_head:               3225942
git_dirty:              0
rows:                   20000
batch:                  500
rounds:                 3000
query_rounds:           200
samples:                3
warmup:                 30
skip_load:              0
----------------------------------------
select1_avg_ms:         0.2678
select1_stdev_ms:       0.0048
raw_load_sec:           1.633
raw_load_rows_per_sec:  12247.4
build_ik_all_sec:       19.799
build_ik_content_sec:   15.633
build_beng_en_sec:      8.332
build_total_sec:        43.782
----------------------------------------
tokenize_ik_avg_ms:     0.7871
tokenize_ik_median_ms:  0.7866
tokenize_ik_stdev_ms:   0.0089
tokenize_beng_avg_ms:   0.4498
tokenize_beng_median_ms:0.4368
tokenize_beng_stdev_ms: 0.0193
----------------------------------------
query_cn_hits:          8001
query_cn_avg_ms:        3.1315
query_cn_stdev_ms:      0.0089
query_beng_hits:        11000
query_beng_avg_ms:      5.9346
query_beng_stdev_ms:    0.0833
query_mixed_hits:       7332
query_mixed_avg_ms:     4.6886
query_mixed_stdev_ms:   0.0676
query_limit_hits:       20
query_limit_avg_ms:     1.4214
query_limit_stdev_ms:   0.0042
========================================

Workflow run