Use of optional .exclude files for selective exclusion during build

[lizziel]

As part of the coupling between CESM and GEOS-Chem, Thibaud Fritz
(@fritzt) added functionality of optional .exclude files to list files in the
sub-tree that do not need to be compiled. This is useful for coupling the
GEOS-Chem chemistry within CESM since some files contained in the
GEOS-Chem repository, such as the advection modules, are not used in
CESM. Using .exclude enables the GEOS-Chem source folder to be
unmodified from its Github repository.

The .exclude files are read from mkSrcfiles during creation of
Srcfiles. Any directory in Filepath could have a local .exclude
file containing filenames to exclude across its subdirectories.

An example of using .exclude for coupling GEOS-Chem within CESM
is here at CAM PR #484.

This PR supersedes PR #3974. It is currently draft until I I get
to running the tests and updating docs as needed. Feedback welcome!

I looked at CIME documentation in doc/source/ but did not see any docs that need updating for this feature.

Test suite: in progress
Test baseline:
Test namelist changes:
Test status: [bit for bit, roundoff, climate changing]

Fixes #3366

User interface changes?: No

Update gh-pages html (Y/N)?: No

[lizziel]

@jedwards4b, I followed the documentation for running tests at https://github.com/ESMCI/cime/wiki/Testing. Heads up that the page is out-of-date and I had to search for the new location of scripts_regression_tests.py. If I should be running other tests let me know.

[lizziel]

I am getting errors like this when I run the test suite. Is this expected?

Finished CREATE_NEWCASE for test TESTRUNPASS_P1.f19_g16_rx1.A.cheyenne_intel in 0.178603 seconds (FAIL). [COMPLETED 2 of 3]
    Case dir: /glade/scratch/elundgren/scripts_regression_test.20220216_121919/TESTRUNPASS_P1.f19_g16_rx1.A.cheyenne_intel.fake_testing_only_20220216_122448-20220216_122448
    Errors were:
        ERROR: Directory /glade/scratch/elundgren/scripts_regression_test.20220216_121919/TESTRUNPASS_P1.f19_g16_rx1.A.cheyenne_intel.fake_testing_only_20220216_122448-20220216_122448 does not appear to be a valid case directory

[jedwards4b]

Nightly test results are here:
https://my.cdash.org/index.php?project=CIME

Looks like we currently have one failing test.