docs: route change requires client daemon reconnect

Document two lessons from the RTSP-camera-via-subnet-route diagnosis:

1. Existing NetBird clients don't automatically re-apply new route
   config; daemons stuck in old state need explicit restart per
   platform (launchctl kickstart / App reconnect / compose restart).
   Shows up as 'Networks: -' on status output and no entry in the
   system routing table.

2. Some iOS apps (notably certain RTSP players) don't use the system
   VPN route and bypass NetBird entirely even when route is
   correctly installed. Safari test distinguishes 'route broken'
   from 'app doesn't respect VPN'.

SUMMARY.md

@@ -88,6 +88,8 @@ reopen fd 8: permission denied
 | Mac `netbird status` 卡在 `Management: Disconnected, rpc error` | daemon 本身卡死，`netbird up`/`down` 救不回 | `sudo launchctl kickstart -k system/netbird`（服務名是 `netbird` 而非 `io.netbird.client`）；見 [peer-deployment-ops.md](./peer-deployment-ops.md) |
 | LXC CT100 SSH root 密碼被擋（`Permission denied`） | sshd_config 內有 `PermitRootLogin prohibit-password` 比 `yes` 先出現，first-match-wins | 加 `/etc/ssh/sshd_config.d/99-allow-root.conf` 覆蓋；socket activation 不需重啟 sshd |
 | 新 peer `netbird status` 顯示 Connected 但互 ping 都不通 | 主機同時跑 Tailscale kernel-mode，`iptables-legacy ts-input` 把 `100.64.0.0/10` 非 tailscale0 封包全 drop，NetBird `100.71.0.0/16` 中彈 | `iptables-legacy -I INPUT 1 -i wt0 -j ACCEPT` + FORWARD 同理；systemd 持久化；詳見 [peer-deployment-ops.md](./peer-deployment-ops.md) |
+| 加 subnet route 後部分裝置拿不到路由（`Networks: -`、`Status: Connecting`） | 既有 NetBird daemon 不會自動重新套用 route，卡在舊 state | 各裝置重連 daemon：macOS `launchctl kickstart`、iOS App Disconnect→Connect、Docker `compose restart`；見 [network-routes.md](./network-routes.md) |
+| iPhone RTSP app 連不到 LAN 攝影機（Safari 打 HTTP 可以） | 某些 iOS app 不走系統 VPN 路由 | 換支援 VPN 的 app（VLC 等），或改用 NetBird FQDN |
 
 ## 檔案位置