This is the SlapOS part.
ERP5 SQL methods must be changed to not hardcode 'InnoDB'.

/reviewed-on nexedi/slapos!368

the recipe slapos.cookbook:generate.password will install the password only after the [request-xx] was called for the first time. Then the second run of buildout the password can change as it was not stored yet. We use publish-early to make it persistent on master and avoid inconsistency between published password and stored one.

/reviewed-on nexedi/slapos!370

This option is not advertised and it is not needed at all in Caddy
configuration.

As monitor stack uses values directly from buildout convert them to buildout
safe values, otherwise the the magic !py! is not stripped.

Also add a test proving that monitor with default values is going to be
correctly configured.

It sometimes results with returning passwords with newlines, thus they have
to be trimmed.

Features:

 * jinja2 is used to generate instance templates
 * downloads are done the same way for all resources
 * create with shared content for all instance profiles
 * fill in instance-common with shared sections
 * render templates late in order to ease its extenension and development
 * drop not needd duplicated section
 * drop slap-parameter in frontend and replicate template
 * simplify monitor configuration
 * move instance-parameter to instance file
   Thanks to this only one and topmost profile is reponsible for parsing and
   passing through the information which comes from the network

make sure password section is installed before request runner{0,1} and pbs instance for webrunner because the password is send in the request.

The path of software release in webrunner is like: `/srv/slapgrid/slappartXX/srv//runner/project/slapos/software/soft/software.cfg`  because ~/srv directory in webrunner is set with a / at the end:

Changing that path will modify the URL of software release in webrunner, this is why we should always keep it as: `/srv/slapgrid/slappartXX/srv//`

if backup.signature is present in tmp/ folder (during push) promise will fail because find return 2 elements and 1 is expected

If pull is running, files can be changed and promise will raise. Wait when backup pull
is not running before check signature diff.

Strip earlier, as stripping so late leads to correct self.passwd
but incorrect passwd, thus 3rd party applications storing passwords
with additional newlines are still resulting in bad passwords.

/reviewed-on nexedi/slapos!366

This is a follow-up of commit 30c5367a

We still don't depend on neoppod[storage-importer,tests]
to avoid pulling ZODB3.

This also fixes pinning issue for software-zodb3.cfg

The backend url can come from request in `url` and `https-url` strings.
It is validated using real caddy template configuration and by using
caddy's `-stdin`. It results with calling it on each slave having any of
those parameters.

This reverts commit 1c9075a1.

The changes will be done on slapos.toolbox/runner to always create missing proxy.db file

add PyWavelets, decorator, networkx
version up: Pillow 5.2.0

Add osd trainned data

This is required starting erp5.utils 0.4.51

Also add a missing md5sum for instance.cfg.in

/reviewed-on nexedi/slapos!362

Fixes:

 * skip tests for parts which won't be fixed
 * fix assertion for master partition
 * fix typos
 * parameterize headers depending on frontend type