Bug when temperature plotting with borefield with different borehole lengths

[wouterpeere]

Currently, GHEtool assumes in a couple of methods that the depth of the borefield is equal to the depth of the first borehole. This can be seen in:
self.H = self.borefield[0]

This is not a problem in sizing, but it is in temperature plotting, for there, the power is divided based on the number of boreholes and the borehole length.

The solution can be to redefine the self.H as being the average of all the borehole lengths when loading a borefield.

[wouterpeere]

Bug is fixed, but one test is now way slower than before (optimise_load_profile_energy). This has to be investigated further.

[wouterpeere]

The timing issue seems to be non-existent.
There is however a deeper challenge w.r.t. the g-function class. When a borefield with two boreholes of 100 and 150m are calculated, the average depth is 125m and it is hence stored in the gfunction class. When later the borefield is changed to have two times 125m, the gfunctions are the same, which is wrong.
This will be solved by setting the 'store_previous_values' to false whenever the boreholes are not the same.