Speed up find_local_max when given a finite num_peaks

[klaussfreire]

Description

Optimizes _shared.coord.ensure_spacing as used in
feature.peak.find_local_max when given num_peaks

There's no need to process all candidates, which could
far outnumber the requested maximum, since candidates
in find_local_max are sorted by intensity. So only return
the first few matches and ignore the rest.

The resulting speedup can be huge in a number of cases

For reviewers

• Check that the PR title is short, concise, and will make sense 1 year
  later.
• Check that new functions are imported in corresponding __init__.py.
• Check that new features, API changes, and deprecations are mentioned in
  doc/release/release_dev.rst.

[jni]

Cool, thanks @klaussfreire! We've definitely been having performance issues with that function so this is much appreciated. The changes are completely internal (no API changes), which also makes it easy to get it in.

[mkcor]

Nice!