:bug: Operations do not converge

[Sviatose]

Water in the dehydrated stream after first run 4.633867677989811 ppm
Water in the dehydrated stream after second run 5.156132446074598 ppm
Water in the dehydrated stream after 5 run 6.4574259175511175 ppm

See GlycolRigTest2

[Sviatose]

https://github.com/equinor/neqsim/tree/bug--operations-do-not-converge

[EvenSol]

I see for the
makeup_calculator.addInputVariable(flash_gas);

you should only add what is not going as strippping gas

[EvenSol]

I tried to increase accuracy of recycle - see:
#910