fix(jianyu): avoid early API bucket cutoff

[jackwener]

Summary

• stop truncating Jianyu API rows inside per-bucket collection
• collect unique rows across all API buckets before later filtering/slicing
• add a regression test proving later-bucket unique rows survive even when earlier buckets fill the nominal limit

Verification

• npm run test:adapter -- --reporter=verbose clis/jianyu/search.test.ts clis/jianyu/shared/procurement-contract.test.ts clis/jianyu/shared/procurement-detail.test.ts
• npm run typecheck
• npm run build