Forest problem has wrong ideal and nadir values

[juropo]

Forest problem has hard coded ideal and nadir values at the moment. If comparing is set to False, these values are wrong.

To fix this issue and to support different data files in the future, these values should be found through optimization. I'm pretty sure that in this specific problem, you will always be able to find ideals by just optimizing for a single objective and the nadirs are the lowest values found when optimizing for one of the other objectives alone.

[jpajasmaa]

Also noticed minor bug on f_2 objective

DESDEO/desdeo/problem/testproblems.py

Lines 1765 to 1775 in cd000d5

	f_2 = Objective(
	name="Wood stock volume",
	symbol="f_2",
	func=f_2_func,
	maximize=True,
	ideal=ideals["f_2"],
	nadir=nadirs["f_3"],
	objective_type=ObjectiveTypeEnum.analytical,
	is_linear=True,
	is_convex=False, # not checked
	is_twice_differentiable=True

The following part looks like a bug:
ideal=ideals["f_2"],
nadir=nadirs["f_3"],