This does not make the code simpler but a little more general,
anticipating for example time steps different than one day.

In main program unit `eddy_graph` and procedures `dispatch_snapshot`
and `read_snapshot`, variable k now refers to the date index, 0-based,
instead of the date in days since 1950-1-1.

In main program unit `eddy_graph`, add variable d for date. `k_begin`,
`k_end` and `k_end_main_loop` also now refer to date indices. We have
to pass an additional argument `d_init` to procedure
`dispatch_snapshot`.

Note that no modification is necessary in procedures `get_snapshot`,
`recv_snapshot` and `send_snapshot` although the meaning of k in
`get_snapshot` and tag in `recv_snapshot` and `send_snapshot` also
changes.

Also split the tests, Fortran source files, Python files and JSon
files. Keep at the top level the files used in both `Inst_eddies` and
Overlap.

Motivations for the split:

- The top directory contained many files.

- It may be useful to compile only one of the two sub-projects. For
example, only `Inst_eddies` on Ciclad, where MPI 3 is not available.

Add main program `eddy_graph`.

Change order of dummy arguments of `dispatch_snapshot` and overlap.

In procedure `dispatch_snapshot`, rename dummy argument m to rank,
clearer.

Rename program `test_get_snapshot` to `test_get_dispatch_snap` and
make it call `dispatch_snapshot` instead of `send_snapshot`. The only
difference in result is that the program now writes files
`isolated_nodes.txt` and `number_eddies.csv`.

Check in program `test_overlap` that it is run on a single MPI process.

Bug fix in procedure `dispatch_snapshot`: the rank of MPI processes is 0-based.

Bug fix: remove `extr_map_outermost.nc`, which was not used since
commit ad420d25.

`settings.mk` in GNUmakefile.

Bug fix in procedure `dispatch_snapshot`: an isolated eddy is a valid eddy.

In script `plot_snapshot.py`, in function snaphot, replace argument ax
with argument `new_figure` so the user does not have to call figure
and `axes(projection = projection)`. Add argument light to function
snapshot so we can call snapshot with light set to False for a second
snapshot. Add a test to only read `h.nc` when we need it.

In script `read_overlap.py`, print more information: number of nodes,
number of edges, number of nodes with at least one successor, number
of nodes with at least one predecessor, splitting events, merging
events.

In script `stat.py`, use the convention that shapefiles are grouped in
a directory instead of identifying a set of shapefiles by the end of
their basename.

In main program `test_successive_overlap`, we know that `delta_in` for
snapshot 1 and `delta_out` for snapshot 2 are `huge(0)` so do not test
this. Print eddy identifiers in a single line. Add the printing of
identifiers of isolated eddies.

extraction of eddies and overlapping.

Procedure dispatch_snapshot will be called by the overlap program and
will not write eddies. So remove the calls to write_eddy from
dispatch_snapshot and move them to test_get_snapshot and
test_read_snapshot.

Do not create files isolated_nodes_1.csv and number_eddies_1.csv in
test_get_snapshot and test_read_snapshot, that will be done by the
overlap program.

init_shapefiles. This allows us to have variables for DBF field
numbers, as module variables. We avoid the burden of passing this
large number of variables as arguments to write_eddy. Safer to have
variables for DBF field numbers than relying on the order of those
fields.

Remove file_list. This was a duplication of information.

Use named constant for missing max speed value. It is used in two procedures.

src_test_get_snapshot.

Move field speed from DBF file max_speed_contour to DBF file extremum
because it can be associated to either the contour in
outermost_contour or in max_speed_contour.

of local variable radius in procedure set_max_speed. Add corresponding
column in outermost_contour DBF file. This is just a diagnostic of the
workings of get_snapshot. Clearer to have this output than having
special meaning in components ssh and area of speed_cont.

program behavior.

In plot_snapshot.py, no need to accumulate all content of shapefile
extremum_1, just process it record by record, as for other shapefiles.

low-amplitude threshold, min_amp.

Add component suff_amp to derived type eddy. Write this component in
extremum_$m.dbf.

In shapefiles, write positions of vertices in degrees rather than
radians.

test_get_snapshot reads the whole domain of input NetCDF files instead
of selecting a window through a namelist. This is clearer and more
convenient. Same in test_get_snapshot.py, test_local_extrema and
test_local_extrema.py.

In test_get_snapshot.py, annotate extrema instead of inserting a
legend.

More correct names: radian_per_degree to deg_over_rad,
degree_per_radian to rad_over_deg.