define a clean and efficient way to handle floating point precision errors in the model

[AteZenmo]

In the model, a lot of places there are hard checks on negative values (sometimes positive as well). Sometimes due to floating point precision errors, these are breached by numbers like -1e-18. This is not an error in our model/code itself. Checks like these are mostly solved now by manual writing a check at that location like if(x < 0 && x > -0.0000001) x = 0; Often the amount of zeros varies at the different places it is used.
This not very clean, and it would be better if we created a default way to handle this, so it works the same everywhere. This should become a function or whatever in the J_GlobalParameters or something like that class in the engine, so it can be accessed everywhere in the model using J_GlobalParameters.precisionCheck(x) or something.

-> Find the best solution for this, and implement it in the model at all the seperate locations it is done now.