Make the force restart from the Watchdog MonitoringThread timeout more obvious.

Currently this event is logged as a WARN log level event:
```json
{
"metadata": {
"device": {
"deviceId": "3d66c50108e4fc01",
"name": "REV Robotics Control Hub v1.0",
"serialNumber": "3d66c50108e4fc01",
"isOnline": true,
"release": "7.1.2",
"sdk": 25,
"featureLevel": 25,
"model": "Control Hub v1.0",
"type": "HANDHELD",
"isEmulator": false
},
"filter": "package:mine ",
"projectApplicationIds": [
"com.qualcomm.ftcrobotcontroller.test",
"com.qualcomm.ftcrobotcontroller"
]
},
"logcatMessages": [
{
"header": {
"logLevel": "WARN",
"pid": 1001,
"tid": 1075,
"applicationId": "org.firstinspires.ftc.ftcaccesspointservice",
"processName": "org.firstinspires.ftc.ftcaccesspointservice",
"tag": "FtcAccessPointService",
"timestamp": {
"seconds": 75,
"nanos": 929000000
}
},
"message": "MonitoringThread: The RC has not reported itself as alive for more than 10 seconds. Relaunching the RC."
}
]
}
```

Displayed as this in Android Studio's logcat viewer:
```logcat
1970-01-01 10:03:40.001 1001-1075 FtcAccessPointService org...res.ftc.ftcaccesspointservice W MonitoringThread: The RC has not reported itself as alive for more than 20 seconds. Relaunching the RC.
```

The warn event makes this not stand out as the cause for an app crash and restart, especially if some more intense computation on startup (which I was causing deliberately) cause the app to bootloop.

Instead a switching to the ERROR log level, or having something spit out accompanying ERROR level messages that make the cause of the restart more obvious to a reader.