Better pareto calculation

[DanieleOranges]

Types of changes

• Docs change / refactoring / dependency upgrade
• [X ] Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)

Motivation and Context / Related issue

The original implementation for dominated points filtering was inefficient, so it get very slow very quickly with pareto size.
In our problems, with a budget of 30k samples, the original implementation needed 2 h runtime, vs 15 min only with the new implementation.

The new implementation comes from https://stackoverflow.com/questions/32791911/fast-calculation-of-pareto-front-in-python

How Has This Been Tested (if it applies)

Some timings on generated numbers

Checklist

• The documentation is up-to-date with the changes I made.
• I have read the CONTRIBUTING document and completed the CLA (see CLA).
• All tests passed, and additional code has been covered with new tests.

[teytaud]

Waow!!! This looks super exciting.
I run the checks.

[teytaud]

This is not a lot of lines, but it is very impactful. Thanks a lot @DanieleOranges !