Fix compiler detection to avoid bad mpicc match

[gsjaardema]

Redo of #956. The previous test would match the MPI mpicc with the test for matching the intel compiler icc. This test uses CMake determined CMAKE_C_COMPILER_ID which is a safer test.

Also fixes a test for the host machine. APPLE was missing CMAKE_HOST_APPLE

[gsjaardema]

Sorry, should have checked before assuming, but APPLE is a valid CMake variable, so does not need to be changed to CMAKE_HOST_APPLE. I can remove that change if desired.

[nmoinvaz]

It does not need APPLE at all because UNIX implies APPLE. So it would be better if APPLE was removed.

[nmoinvaz]

LGTM thank you!

[gsjaardema]

I think a maintainer has to click a button to allow the workflows to run for this PR. I am a first-time contributor to this repository.

[nmoinvaz]

@Dead2 will need to approve.

[Dead2]

Did anyone actually test this with an Intel compiler to verify we don't have any regressions?

[codecov]

Codecov Report

Merging #959 (0ea7b59) into develop (54b1c13) will decrease coverage by 0.01%.
The diff coverage is n/a.

@@             Coverage Diff             @@
##           develop     #959      +/-   ##
===========================================
- Coverage    77.36%   77.34%   -0.02%     
===========================================
  Files           74       74              
  Lines         8309     8309              
  Branches      1374     1374              
===========================================
- Hits          6428     6427       -1     
  Misses        1348     1348              
- Partials       533      534       +1     

Flag	Coverage Δ
macos_clang	68.55% <ø> (ø)
macos_gcc	67.45% <ø> (ø)
ubuntu_clang	73.61% <ø> (ø)
ubuntu_clang_debug	73.65% <ø> (ø)
ubuntu_clang_inflate_allow_invalid_dist	73.34% <ø> (ø)
ubuntu_clang_inflate_strict	73.60% <ø> (ø)
ubuntu_clang_mmap	73.31% <ø> (ø)
ubuntu_clang_msan	73.61% <ø> (ø)
ubuntu_gcc	73.00% <ø> (ø)
ubuntu_gcc_aarch64	74.48% <ø> (-0.02%)	⬇️
ubuntu_gcc_aarch64_compat_no_opt	73.32% <ø> (ø)
ubuntu_gcc_aarch64_no_acle	73.62% <ø> (-0.03%)	⬇️
ubuntu_gcc_aarch64_no_neon	74.08% <ø> (ø)
ubuntu_gcc_armhf	74.48% <ø> (-0.02%)	⬇️
ubuntu_gcc_armhf_compat_no_opt	73.31% <ø> (ø)
ubuntu_gcc_armhf_no_acle	74.67% <ø> (-0.02%)	⬇️
ubuntu_gcc_armhf_no_neon	75.07% <ø> (ø)
ubuntu_gcc_armsf	74.49% <ø> (-0.02%)	⬇️
ubuntu_gcc_armsf_compat_no_opt	73.31% <ø> (ø)
ubuntu_gcc_compat_no_opt	74.42% <ø> (ø)
ubuntu_gcc_mingw_i686	0.00% <ø> (ø)
ubuntu_gcc_mingw_x86_64	0.00% <ø> (ø)
ubuntu_gcc_no_avx2	74.52% <ø> (ø)
ubuntu_gcc_no_pclmulqdq	71.46% <ø> (ø)
ubuntu_gcc_no_sse2	72.54% <ø> (ø)
ubuntu_gcc_no_sse4	71.42% <ø> (ø)
ubuntu_gcc_o3	73.79% <ø> (ø)
ubuntu_gcc_osb	75.98% <ø> (ø)
ubuntu_gcc_ppc	74.80% <ø> (ø)
ubuntu_gcc_ppc64	75.56% <ø> (ø)
ubuntu_gcc_ppc64le	74.32% <ø> (ø)
ubuntu_gcc_s390x	73.23% <ø> (ø)
ubuntu_gcc_sparc64	76.03% <ø> (ø)
win64_gcc	70.39% <ø> (ø)
win64_gcc_compat_no_opt	?

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
arch/arm/adler32_neon.c	97.50% <0.00%> (-2.50%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 54b1c13...0ea7b59. Read the comment docs.

[gsjaardema]

@Dead2 Yes, I have tested it with CC=icc cmake .. and with CC=mpicc cmake .. (where mpicc is based on an intel compiler) and both correctly identify that the compiler (or underlying compiler) is intel-based. Configuration suceeds; build succeeds and tests complete.

[scivision]

I tested this on Windows as well. Before this PR, GCC mpicc fails, and with this PR, it's fixed. Thank you!