Optimize computation of Leader Selection during EFM

[AlexHentschel]

PR comment suggesting dynamic leader extension

I think there might be value in consolidating the logic for the leader selections, purely for readability and encapsulation (reducing surface for possible bugs).

Def of done

provide an extendable leader selection, that in addition memorizes the rng and has a slice of extensions, where each extension is its own leader selection

• we already have a binarySearchStrictlyBigger method that can be copied and extended to work on the extension and find the leader-selection, whose range contains the requested view
• in documentation of binarySearchStrictlyBigger replace "non-decreasing" by "monotonically increasing" (for every indices i,j with i<j we must have arr[i] ≤ arr[j])

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.