Issue warning for fluid sensors with one port

[mwetter]

This issue is to remove sensors in IBPSA.Fluid.Sensors that have only one port and whose output depends on the flow direction.

Instead of these sensors, it is better to use their version that have two fluid ports, as is described in IBPSA.Fluid.Sensors.UsersGuide.

The one port sensors are removed because many users use them, including users who attended a workshop where we explained to them when and why not to use them, and made them aware of the info in IBPSA.Fluid.Sensors.UsersGuide.

See also https://github.com/ibpsa/project1/wiki/2020-09-15-library-development#1-port-sensors-such-as-for-temperature

[mwetter]

Todo:

• Add warnings to sensors that have one port.
• Update user guide.
• Update conversion script.

[mwetter]

There are a few use cases where the sensors with one port make sense, and the use of sensors with two ports would be non-intuitive, or can lead to wrong results if the sensor is configured as dynamic. For example, the image below shows how to measure PPM from a volume with a two port sensor. If configured to be dynamic, then the result is wrong (because there is no flow through the sensor, its output never changes).

So, a user would need to configure it as steady-state to measure the PPM in the volume, and likely also set allowFlowReversal=false, although in Dymola, the inflowing enthalpy at port b is equal to the outflowing enthaly if unconnected. (I don't think the standard defines the behavior for this case.)

In view of this, it may be better to write a warning if these sensors are used, and allow the user to disable the warning through a parameter. As there are warnings issued, I think it is fine the leave the temperature sensor with one port in the Fluid.Sensors package. If we were to move it to the Obsolete package, users may use the sensor from the MSL, and then don't get a warning. So I think it is better to leave the sensor so users are warned about potential problems.