Add benchmark methodology and pricing research after document migration

scripts/audit-ai-research.py

@@ -55,6 +55,26 @@ def source_register():
 
 
 
+def context_methods():
+    evidence = {e['id']: e for e in DATA['evidence']}
+    rows = [s for s in SOURCES['sources'] if s.get('methodology')]
+    assert {s['id'] for s in rows} == {'S06', 'S07', 'S08', 'S09'}
+    lines = ['', '## 背景基准适用性与缺项', '',
+        '以下核对背景资料的方法及适用范围，不是公司名录的完整覆盖。原始方法记录与全局核查随研究版本进入同一数据库；没有把行业汇总值新增为任何公司的收入观测。', '',
+        '| 来源与证据 | 样本与统计单位 | 时间 | 本次读取范围 | 可用于什么 | 主要限制 |',
+        '| --- | --- | --- | --- | --- | --- |']
+    for source in rows:
+        m = source['methodology']
+        assert m['checkedAt'] <= DATA['asOf']
+        assert all(m.get(key) for key in ('sample', 'unit', 'period', 'access', 'use', 'limit', 'evidenceIds'))
+        assert all(eid in evidence and evidence[eid]['companyId'] is None for eid in m['evidenceIds'])
+        assert not any(o['evidence'] in m['evidenceIds'] for o in DATA['observations'])
+        proof = '、'.join(f"[{eid}]({evidence[eid]['url']})" for eid in m['evidenceIds'])
+        lines.append(f"| {source['id']} {source['name']}：{proof} | {m['sample']}；{m['unit']} | {m['period']} | {m['access']} | {m['use']} | {m['limit']} |")
+    lines += ['', '本表的应用解释属于研究判断。选样、分母、时间或指标不同会造成表观差异；这些差异不单独证明财务造假，也不能作为给未知公司填补收入的依据。', '']
+    return lines
+
+
 def candidate_register():
     batches = DATA.get('candidateBatches', [])
     companies = {c['id']: c for c in DATA['companies']}
@@ -199,6 +219,7 @@ def render():
     lines.append(f"| 其他官网、评测及公开资料 | {len(others)} | {len({e['url'] for e in others})} | 按每条来源限制理解，不能统一当作独立证据 |")
     lines += directory_rechecks()
     lines += source_register()
+    lines += context_methods()
     lines += candidate_register()
     lines += ['', '## 尚未满足的研究条件', '',
         '行业ARR比较仍缺统一业务截止日、币种、经常性定义和业务边界。现有平台MRR可描述已采集产品，不足以代表全部新公司。技术评测只覆盖所列版本与任务，不宜跨领域汇总为统一强弱分数。', '',