Monitoring and Safety

Apart from the ability to do a _small_ experiment we do need monitoring and related to that, safety systems.

Some information should be monitored all the time, and potentially automatically be added to a given experiment, i.e. _Fridge temperature_

As an example, We are running these fridges, I want an service running in the background and ask the fridge for its information every second, or five.

That service should at the same time allow the experiment to talk to the fridge.

In this case the monitor should be there for something like this:
- Log the temperature.
- There is no reason to ask for the fridge temperature 100 times a second during a measurement, but the data could be saved with timestamps that allows to correlate the data.
- Or the measurement asks the service about the latest temperature without interrupting the fridge.
- Other Instruments, in the case here the Magnet should monitor this service as well.
- I certainly want to comply to safety measures, to enforce magnet use only at certain conditions.
There should be global flags that the magnet cheks to know if it is allowed to ramp the field.
- I also want to handle events, The safety monitor should even abort an experiment and keep the system save.

`Temperature > xK: Ramp to zero`,

`Temperature also > yK: Ramp to zero with this slow/fast rate`

`Time == 8PM: order pizza`
- Monitor the status of the electricity grid, this building here is really bad in terms of grid stability
- Notify me on the events I specify

I can think of all kinds of ways to extend this :)
But there is at least one crucial basic features that we need - safety.

The issue names no files, tests, or entry points; start by clarifying the monitoring, safety, event, notification, and instrument-control requirements before locating the relevant experiment interfaces. Done would require an agreed scope for background polling, timestamped data, safety flags, emergency actions, and notifications.

Written by the indexing model from the issue text.