Suse SLE 15 SP6 has no official ansible package.
Instead, we use the official Ansible installation method, by installing
it using pip.

See merge request nexedi/slapos.core!655

The `lsb_release` command is required in order to run the playbook,
because it uses `ansible_lsb`.
Otherwise, the following error is raised:

```
fatal: [127.0.0.1]: FAILED! => {"msg": "The conditional check '(ansible_distribution == \"Ubuntu\" and ansible_distribution_version == \"20.04\") or (ansible_lsb.codename == \"kylin\" and ansible_lsb.major_release == \"v10\")' failed. The error was: error while evaluating conditional ((ansible_distribution == \"Ubuntu\" and ansible_distribution_version == \"20.04\") or (ansible_lsb.codename == \"kylin\" and ansible_lsb.major_release == \"v10\")): 'dict object' has no attribute 'codename'. 'dict object' has no attribute 'codename'\n\nThe error appears to be in '/tmp/tmpplaybook_unstableslapos.27149/roles/repository/tasks/main.yml': line 91, column 5, but may\nbe elsewhere in the file depending on the exact syntax problem.\n\nThe offending line appears to be:\n\n\n  - name: Install Ubuntu 20.04 repository key\n    ^ here\n"}
```

As the package containing `lsb_release` is different on different Linux
distros, this commit only attempts to install it in SLE 16 SP6, where
it is called `lsb-release`.

See merge request nexedi/slapos.core!655

   rely on portal_template API to compareVersions

   Create a ticket if the SlapOS version into the monitored compute node is lower them minimal supported version found for the current master.

  If you access the listbox before the subscription request is indexed the link is broken on the listbox.

  patch provided by: Romain Courteaud.

It prevents Unauthorized, as only owner can access the temp object

  The code was rework to be visible by anonymous

the vulnerability database content seems to have changed

Stop and delete b2b services

- Support Slave Instance allocated in the same tree
- allow creating allocation consistency ticket on Instance Node
- handle allocation consistency for Slave Instance without Instance Node

This is a fixup of d42de18b.

history and statistics files are sometimes corrupted because of invalid json

We do 2 things to improve this:
 1. we use json.load to read the previous file (this will detect if the file is corrupted)
 2. we use json.dump to write the entire file at once in a temporary file and then we move this file to be atomic.

Probably this will consume more resources.

See merge request nexedi/slapos.core!728

  This prevents installation update modification date due Web Page
  modification date is newer.

  This fixes the coding style tests broken by:

    slapos_web_deploy: Added missing standalone-test
    slapos_web_deploy: Add rplc_coupler deploy script

  It was create a while ago, but never pushed.

Do not acquire from Person to an Organisation, as most user do not have access to the organisation_module

- directly show the software_release and sofware_type informations
- instance title is meaningless on this view, as admin want to get an understanding of which kind of software are instanciated
- drop creation_date, image columns