Correct build_r (currently SegFault for alpha=2)

[Philou31]

[mknaranja]

@julianlitz Did #60 actually solve this error or is this still persisting? Could you check in case you don't know?

[julianlitz]

Yes it did resolve this error.
I implemented this fix in the refactored code as well and so far I didn't notice any issues.
But we should check for more r_jump values than the 4 cases to be sure.
Especially if r_jump is close to a boundary value.

However I find the value of r_jump \in (0, 1) to be very unintuitive for the user.
Since „r_“ appears in the variable name, one suspects it corresponds to a specific radius, but it rather seems to be a percentage of where to find the actual alpha_jump \in (0, Rmax).

I believe it would make more sense to pass in a alpha_jump radius instead of a percentage.