feat(overview): graph_overview 全量输出 finding_list

findings 是任务最高价值产物且单任务通常不多，改为在 graph_overview
里全量带出 finding_list（不像 facts 只给最近窗口），planner/worker 每轮
即可一眼看全所有确认漏洞，无需再调 list_findings。

每条精简为 {id, summary, evidence?, from_intent?, assets?}：
- evidence 取 finding payload 的 evidence.poc
- from_intent 为产生该漏洞的意图（yields 边）
- assets 直接给受影响资产的可读内容（url/域名/ip:port 等，不再是裸 id），
  经 FindingMetaByNodeID 反查锚定资产、批量 GetByIDs 解析
- vulnclass/severity/state 等仍走 list_findings / node_detail

新增 AssetStore.FindingMetaByNodeID 薄封装，让只持有 AssetStore 的
ToolSet 能拿到 node_id→资产映射（读 anchors 表，覆盖新旧漏洞）。

db/findings.go

@@ -742,6 +742,13 @@ type FindingMeta struct {
 	AssetIDs []int64
 }
 
+// FindingMetaByNodeID maps a task's finding node ids to their standalone-row
+// metadata (status + anchored asset ids) via the asset store, so callers holding
+// only an AssetStore (e.g. the agent ToolSet) can reach it without a raw *DB.
+func (a *AssetStore) FindingMetaByNodeID(taskID int64) (map[int64]FindingMeta, error) {
+	return a.db.FindingMetaByNodeID(taskID)
+}
+
 // FindingMetaByNodeID maps a task's finding node ids to their standalone-row
 // metadata, so the per-task view (which reads exploration nodes) can show and
 // edit the same status — and the same anchored assets — as the global 发现 page.