For reasons unknown to anyone but the geniuses from Redmond: In
MSWindows, the wildcard pattern "*.gie" expands to all files having
an extension *starting* with ".gie".

In other words "dir *.gie" will list files with extension ".gie"
*and* with extension ".gie_failing".

This means that running "gie test\gigs\*.gie" will run all gigs
tests, even the ones known to fail.

So rather than getting a quick check for regressions, one gets
a lot of noise and no easy way to be sure whether any regressions
have appeared.

To work around this oddity, we rename all files ending in
".gie_failing" to end in ".gie.failing" - and while at it,
adding all the non-failing gigs tests to the AppVeyor test suite.

This extends the regression test suite, run at the end of each
AppVeyor integration, to over 2500 individual tests.

Test coverage, however, is still hovering just below 75%, so
there's still plenty of room and reason for additional tests.

All the GIGS JSON files have been converted to the gie format
and are now covered by running "make check". Therefore we no
longer need the Python setup. Removing it from the Travis CI
setup should speed up the test execution time a bit.

Make sure that the gie files in test/gie and test/gigs are included
in the distribution tarball. Tests with gie are set up to run all
available files matching *.gie in the gie and gigs directories. For
this reason tests that are known to fail in gigs/ have been renamed
to *.gie_failing, so that "make check" can be run in-tree in local
copies of the git repository.

The GIGS tests that are known to work are added to the CMake test
setup. The GIGS gie files have been auto-translated from the
existing json-files and some corrections to tolerances have been
necessary since gie uses different norms than GIGS specify. The GIGS
tolerances are specified as the infinity norm of angular coordinates,
whereas gie uses the actual distances between calculated and expected
coordinates (using geodesics). In a few tests +towgs84 is overriden
from the EPSG inits to avoid creeping numerical inaccuracy in
roundtrips.

Co-authored-by: Thomas Knudsen <thokn@sdfe.dk>
Co-authored-by: Kristian Evers <kristianevers@gmail.com>