[Task] Please increase the number of retries to reduce the frequency of problems

### Description

问题：因为网络原因，一些主动监控，只有一次超时就会告警
建议：麻烦学习一下uptime-kuma增加重新尝试次数和尝试间隔，比如尝试3次失败后再报问题
风险：当然增加重试次数会增加服务负载，这块也可以交于使用者来掌控

### Task List

--》增加重试次数
--》增加重试间隔时间
--》增加状态变化时才告警通知，比如上次状态是不可用，本次检查状态仍未不可用就不用告警，除非状态不匹配才发送告警

No files, tests, or entry points are identified. Start by locating the active-monitor check and alert-notification paths, then clarify how retry count, retry interval, and user control should work. Done means transient timeouts are retried before alerting and repeated alerts are suppressed unless the monitored state changes.

Written by the indexing model from the issue text.