Because it contains timing.

Useless in this script.

Now that we read SHPC on the command line and the SHPC are not
necessarily separated by year, we can use input other than
`$PWD/Global_1993_2020`, and we do not have to specially prepare a file
`sgemented.gt` for this test.

Simply printing g displays the address of the graph, which may change
between runs. This was not convenient for the comparison of runs.

To ease comparison of runs.

Do not assume that each SHPC corresponds to a given year. Do not
impose any directory structure on the input SHPC. We have to bisect
through the list of `d_init` to find the SHPC containing a given date,
as in program `eddy_graph`.

Revert commit 8a1f1d1e for `eddy_graph.sh.in` because timing is not
convenient for automatic comparison of runs.

Add a test that is going through `is_node_type` with an integer label.

Since we no longer load all the contents of shapefiles for a given
year in advance, we no longer need to process the segments year by
year. Motivation: this will allow us to get free of the imposed
arborescence of SHPC by year.

Create properties in `cost_function.py` rather than in
`segments.py`. We do not use these properties in `segments.py`.

The property in the input edgelist is not the non-local cost function.

This test requires a specially prepared file `segmented.gt`:

```
head -n 100000 /data/guez/Oceanic_eddies/Global_1993_2020/Eddy_graph_anti/edgelist.csv >edgelist.csv
~/Detection_eddies/segments.py edgelist.csv segmented.gt
```

because we need input with `%Y/SHPC_(anti|cyclo)` arborescence for the
test.

Simply printing g displays the address of the graph, which may change
between runs. This was not convenient for the comparison of runs.

Motivation: being able to choose a different output format at
run-time. We use the graphml format for the automated test, because
graphml is a text format, so we can compare the results of two runs.