Fixup small kernel situation

[jgfouca]

It was a bit surprising that only the p3_main test actually did something different with small kernels turned on. After reviewing the code, I think I understand why. All the _disp functions are just thin wrappers around the corresponding p3 function. I don't think it's really worth the effort to have separate BFB unit tests for all the _disp functions. The p3_main test (which does test the small kernel stuff) should be adequate.

I've added some comments and change the p3_sk test to only test p3_main since any other bfb testing would be redundant.

[mergify]

Merge Protections

Your pull request matches the following merge protections and will not be merged until they are valid.

🟠 Enforce checks passing

Waiting checks: gcc-cuda / ${{ matrix.build_type }}, gcc-cuda / dbg, gcc-cuda / opt, gcc-cuda / sp, gcc-openmp / ${{ matrix.build_type }}, gcc-openmp / dbg, gcc-openmp / opt, gcc-openmp / sp.

Make sure that checks are not failing on the PR, and reviewers approved

• any of:
  • check-skipped={% raw %}gcc-openmp / ${{ matrix.build_type }}{% endraw %}
  • all of:
    • check-success="gcc-openmp / dbg"
    • check-success="gcc-openmp / opt"
    • check-success="gcc-openmp / sp"
    • check-success="gcc-openmp / fpe"
• any of:
  • check-skipped={% raw %}gcc-cuda / ${{ matrix.build_type }}{% endraw %}
  • all of:
    • check-success="gcc-cuda / dbg"
    • check-success="gcc-cuda / opt"
    • check-success="gcc-cuda / sp"
• #approved-reviews-by >= 1
• #changes-requested-reviews-by == 0
• any of:
  • all of:
    • check-success="cpu-gcc / ERS_Ln22.ne4pg2_ne4pg2.F2010-SCREAMv1.scream-small_kernels--scream-output-preset-5"
    • check-success="cpu-gcc / ERS_Ln9.ne4_ne4.F2000-SCREAMv1-AQP1.scream-output-preset-2"
    • check-success="cpu-gcc / ERS_P16_Ln22.ne30pg2_ne30pg2.FIOP-SCREAMv1-DP.scream-dpxx-arm97"
    • check-success="cpu-gcc / SMS_D_Ln5.ne4pg2_oQU480.F2010-SCREAMv1-MPASSI.scream-mam4xx-all_mam4xx_procs"
  • check-skipped={% raw %}cpu-gcc / ${{ matrix.test.short_name }}{% endraw %}
• any of:
  • check-skipped=cpu-gcc
  • check-success=cpu-gcc

[bartgol]

I'm going to merge this, since testing will continue to fail until baselines are regenerated