Changed heat consumer flow control when qext_w is zero

[JonasPfeiffer123]

In reference to Issue #679 regarding time series calculation problems, I tested setting the mass flow in the heat consumer to zero instead of deactivating the component. This approach seems to resolve the issue I encountered. What are your thoughts on this change? Do you see alternative solutions?

Additionally, I face two further challenges:

• How should the network handle scenarios where all heat consumers have a qext_w of zero in a time step?
• During time steps with minimal overall load, the temperature drops below the heat consumers' return temperature, causing pipeflow to fail due to cooling in the pipes. Increasing text_k in the pipes mitigates the problem, but the results become unrealistic since the ground temperature is not accurately reflected.

What would you recommend to address these challenges?