Inconsistency between the MFS of the BruteForce and BlackBox

**Describe the bug** Our implementation of brute force and black box algorithms for the Minimal Fatal Shock are inconsistent. The BlackBox variant returns a vector that leads to a different attractor (as defined) while the Brute Force variant returns a vector that leads to the same attractor (NOT as defined). This is wrong and it is simply because of the logic of the for loop: the brute force version must return the second-to-last estimated vector, not the last one, as the last one leads to the same basin. **Minimal Working Example** One can use the example of the documentation and swap the two algorithms and then call the output with the given `mapper`.