feat: initialize bangwen recruitment aggregation and search engine

- Rust core engine with HTTP ingestion gateway, hybrid search, and white-box resume matching
- Python edge collectors supporting declarative TOML and specialized adapters
- PostgreSQL schema migrations with pgvector embedding support
- Hardened security controls: strict CSP, terminal sanitization, and input boundary validation
- Multi-service container orchestration with dedicated non-root execution

migrations/20261023000003_data_cleanup_and_limits.sql

@@ -0,0 +1,26 @@
+-- Migration: Clean up excessive attachments, dirty bodies, and orphan test sources
+
+-- 1. Cap excessive historical attachments per announcement (keep newest 50 per announcement)
+DELETE FROM attachment
+WHERE id IN (
+    SELECT id FROM (
+        SELECT id, ROW_NUMBER() OVER (PARTITION BY announcement_id ORDER BY id DESC) AS rn
+        FROM attachment
+    ) t
+    WHERE t.rn > 50
+)
+AND id NOT IN (SELECT attachment_id FROM "position" WHERE attachment_id IS NOT NULL);
+
+-- 2. Clean dirty, empty, or anti-crawler body texts and clear associated hashes
+UPDATE announcement 
+SET body_text = NULL, simhash = NULL, content_hash = NULL 
+WHERE body_text = '' 
+   OR (LENGTH(body_text) < 120 AND body_text ~* '\.(pdf|docx?|xlsx?|zip)$')
+   OR LENGTH(body_text) < 60
+   OR body_text LIKE '%环境异常%' 
+   OR body_text LIKE '%完成验证%';
+
+-- 3. Delete orphan test sources with 0 announcements
+DELETE FROM source 
+WHERE id IN ('test_auth', 'yn.ynrsksw.special8')
+  AND NOT EXISTS (SELECT 1 FROM announcement WHERE announcement.source_id = source.id);