WIP: Flatter SetBounds

[bprather]

PR Summary

This moves the t, u and v indices of the kernel in SetBounds to the outer loop from their current spot in the inner loop. This will help to launch enough teams via the outer loop in codes with a single variable and single block per rank, where the total number of boundary buffers on a rank may be quite small.

Because all the indices are abstracted, one could imagine using basically this code to instead specify a desired number of teams, or amount of work per team -- the additional flexibility might let a broader range of codes realize a little speedup.

Remaining todo:

• Add this tweak to SendBoundBufs as well
• Syntactic sugar to avoid the max(outer_product(Nx...)) shenanigans
• Potentially allow more flexible tweaking of outer/inner loop size at runtime (possibly integrates with what @pgrete was doing with loop sizes during the hackathon). Probably useful in many/most nested-parallel loops.

PR Checklist

• Code passes cpplint
• New features are documented.
• Adds a test for any bugs fixed. Adds tests for new features.
• Code is formatted
• Changes are summarized in CHANGELOG.md
• Change is breaking (API, behavior, ...)
  • Change is additionally added to CHANGELOG.md in the breaking section
  • PR is marked as breaking
  • Short summary API changes at the top of the PR (plus optionally with an automated update/fix script)
• CI has been triggered on Darwin for performance regression tests.
• Docs build
• (@lanl.gov employees) Update copyright on changed files

[bprather]

@par-hermes format

[Yurlungur]

@bprather reminder to make the magic happen

[Yurlungur]

@lroberts36 reminder to make the magic happen