开启 backup request 和 retry times 以及自定义 retry policy时会触发 CHECK_EQ 失败

**Describe the bug (描述bug)**
当重试次数>=3时, 开启 backup request 以及 retry policy 时,
会触发 CHECK_EQ(current_id(), info.id) << "error_code=" << _error_code;

**To Reproduce (复现方法)**
1) 发送一次rpc: 记作 RPC_1. 假设在 backup request time 后返回, 且 retry policy 必定为真.
2) 触发 backup request, 记作 RPC_2. 且返回在 RPC_1 之后.
3) RPC_1 返回后, 重试时触发 CHECK_EQ(current_id(), info.id) << "error_code=" << _error_code;

**Expected behavior (期望行为)**
不触发 CHECK_EQ(current_id(), info.id) << "error_code=" << _error_code;

Reproduce the described RPC_1/RPC_2 sequence with backup requests, at least three retries, and a custom retry policy. Trace the backup-request and retry handling around the CHECK_EQ(current_id(), info.id) assertion; done means this sequence no longer triggers the assertion and existing behavior remains covered by tests if available.

Written by the indexing model from the issue text.