Keep the good contours found when looking for the outermost contour,
subject to a maximum of `n_max_cont` contours. The idea is to use them
later when looking for the maximum-speed contour. Inside function
`get_1_outerm`, we temporarily store the good contours in a local
explicit-shape array, `cont_list`.

Add default initialization for component area of derived type
`ssh_contour`, because we are going to keep contours for which we will
not compute area.

Because the numbering changed when we decided not to write invalid eddies.

I do not remember why I chose to write to standard error. First it is
clearer to write only error messages to standard error. Second the
order of messages on standard error coming from the Fortran code and
from the shapelib C library does not seem deterministic, and this is
annoying to compare runs.

Because the numbering changed when we decided not to write invalid eddies.

Because `$tests_old_dir/Extraction_eddies_region_5/SHPC` is opened
with `SHPC_class` in test `Plot_eddy_contours_window`, which will
create `n_slices.txt`, so there will always be this file only in
`$tests_old_dir/Extraction_eddies_region_5/SHPC`.

This is going to be simpler. Also, we will spare main memory when we
have a list of contours for each eddy, if we do not have a list for
invalid eddies.

Use test `Region_5_15_days` instead of outside experiment `Inst_eddies_4`.

We import in the test infrastructure the experiment `Inst_eddies_4`
because we want to no longer keep invalid contours, so we will need to
update the experiment.

Only define and print `n_radius4` and `n_missing_speed` if `with_radius_4`.

We cannot test `"radius4" in shape_rec_outer.record`, as
`shape_rec_outer.record` is not a dictionary. We could test `"radius4"
in shape_rec_outer.record.as_dict()` but it is better to test outside
of the loop in the field names of the shapefile reader object. The bug
comes from commit 83a61016.

Add default initialization of component `coord_proj` of type
extremum. As the type of `coord_proj` is integer, there may not be a
compiler option to initialize it. We want to detect when it has not
been defined.

As `grid_lon_lat` may be false, we define `e%extr%coord_proj`.

I do not see why this was not done in commit 24b5b079.

This shortens the file `test_get_1_outerm.f90`.

We will need this new shapefile in program `eddy_graph` only for a non
longitude-latitude grid. We add a new dummy argument `grid_lon_lat` to
procedure `shpc_create` to decide whether we create the shapefile
`extr_proj`.

Because we are going to use it also for another shapefile than extremum.

It seems simpler than putting it in `$large_input_dir`. I do not see a
drawback.