GitLab

Taken from the Jenkins Vidjil-browser-functional job

As server configuration may change depending on server, the SSH client was
configured directly on the server rather than on the .gitlab-ci.yml file.

Example:
Host vda.vidjil.org
        StrictHostKeyChecking no
        UserKnownHostsFile=/dev/null
        IdentityFile = ~/.ssh/bonsai-ci

According to the documentation <https://docs.gitlab.com/ce/ci/yaml/#stages>:
“Jobs of the same stage are run in parallel”.

Therefore a stage test_unit seems to make sense where all unit tests would be
launched in parallel.

The traceback seems to change depending on the architecture.
This should be more general while still being precise.

See #2282.

Completes 2f117f42 and fixes #2281.

Follows 71949d8b and fixes #2273.

See #2282.
Follows 7db1b5f5 and 7b1a71d1.

If the server was not happy about the input it will complain
by putting an error message in the response.
We display it in such a case.

We also test that behaviour (and that an unsegmented sequence is actually
displayed).

See #2275. See also #2034 for discussion on JSON response formatting.

It can be used more easily afterwards (eg. for #2275).

The feature is not ready on the server side (see #2071)

If the sample is of generic type the model data was not populated,
leading to some confusing information in the model left over from
previous samples

the parameter in the given url did not match the url that was loaded by
the onclick function, and therefore was not working

Follows 71949d8b. See #2273.

'germline' may be a global variable.
Completes 0ca30a56 and fix #2267.

See #2270.

See #2270.

given that the same file is used to fake_results_file 1 and 2, the
actual file for fake_results_file 2 was empty since the stream had not
been reset before attempting to store the file.

with the arrival of the generic sample_set, the sample_set list was
refactored, into a single route for all sample_set types in order to
eliminate some code duplication.