feat(codex): 增强 OAuth 导入解析与账号信息提取，优化维护调度器线程模型

Codex OAuth:
- 导入解析支持附加账号字段（account_id/plan_type/user_id/email）
- enrich_codex 扩展从 access_token 和直接字段提取账号信息
- parse_codex_id_token 支持 JWT/JSON 字符串/dict 三种输入格式
- request patching 新增 openai:compact 格式支持
- codex_usage_parser 新增 credits_unlimited 字段解析

维护调度器:
- 同步 DB 操作迁移到线程池执行，避免阻塞事件循环
- 新增 request_candidates 定期清理任务
- 新增每周 VACUUM ANALYZE 数据库表维护任务
- 新增 enable_db_maintenance 配置项

前端:
- ElapsedTimeText 从 setInterval 改为 requestAnimationFrame
- 时间线过滤 available/unused 占位记录
- Usage 页面默认关闭全局自动刷新
- 移除号池管理中的配额更新时间显示

frontend/src/features/usage/components/HorizontalRequestTimeline.vue

@@ -642,6 +642,10 @@ const normalizeTimelineStatus = (value: unknown): CandidateRecord['status'] => {
   return 'failed'
 }
 
+const isPlannedOnlyStatus = (status: CandidateRecord['status']): boolean => {
+  return status === 'available' || status === 'unused'
+}
+
 // 候选时间线（按实际执行顺序排序）
 const rawTimeline = computed<CandidateRecord[]>(() => {
   if (!trace.value) return []
@@ -658,6 +662,11 @@ const rawTimeline = computed<CandidateRecord[]>(() => {
     })
 })
 
+// 仅保留“已进入执行链路”的节点，过滤预创建但未执行的 available/unused 占位记录。
+const executableTimeline = computed<CandidateRecord[]>(() => {
+  return rawTimeline.value.filter(candidate => !isPlannedOnlyStatus(candidate.status))
+})
+
 const schedulingAudit = computed<Record<string, unknown> | null>(() => {
   const metadata = props.requestMetadata
   if (!metadata || typeof metadata !== 'object' || Array.isArray(metadata)) return null
@@ -677,7 +686,7 @@ const extractPoolGroupId = (candidate: CandidateRecord): string | null => {
 
 const poolAttemptCandidates = computed<CandidateRecord[]>(() => {
   // 新链路：优先使用后端写入的 extra_data.pool_group_id。
-  const fromTrace = rawTimeline.value.filter((candidate) => extractPoolGroupId(candidate) !== null)
+  const fromTrace = executableTimeline.value.filter((candidate) => extractPoolGroupId(candidate) !== null)
   if (fromTrace.length > 0) {
     return fromTrace
   }
@@ -765,8 +774,8 @@ const poolAttemptKeySet = computed<Set<string>>(() => {
 })
 
 const timeline = computed<CandidateRecord[]>(() => {
-  if (poolAttemptCandidates.value.length === 0) return rawTimeline.value
-  return rawTimeline.value.filter(
+  if (poolAttemptCandidates.value.length === 0) return executableTimeline.value
+  return executableTimeline.value.filter(
     (candidate) => !poolAttemptKeySet.value.has(makeAttemptKey(candidate.candidate_index, candidate.retry_index)),
   )
 })