Use the format return code to distinguish between failure to format or
failure to report to master at the end. Then make boot run format even
if the master is unreachable, and:
* if format fails to report, wait until master is reachable, and retry
* if format fails for another reason, retry after a fixed 15s delay
* once format succeeds fully, bang

Partitions no longer need to be restarted at boot since now slapgrid
will start the services even if there is no connection to master.

This reverts commit cada4581.

Supervisord files of stopped partitions (only those) should be removed,
and those of started partitions should exist.

Create the supervisord configuration file for the partition's processes
on start, and remove it on stop.

Give each supervisor process group its own separate configuration file.

See merge request nexedi/slapos.core!519

See merge request nexedi/slapos.core!517

   Ignore to check if price matches if the invoice was created manually, and not linked to a business process.

See merge request nexedi/slapos.core!516

Improve hateoas support to show shared instances.

It turns out that ERP5 returns
```
'form_id': {
  'type': 'StringField',
  'key': 'form_id',
  'default': 'InstanceTree_viewAsHateoas',
},
```
for the `form_id` in hateoas, so adapt the proxy to do the same.

Currently hateoas support in proxy only lists non-shared instances,
so this always says that the instances are not shared.

See merge request !511

See merge request !514

   Set it non-editable

See merge request nexedi/slapos.core!513

See merge request !509

   This script isnt used at all

See merge request !504