This is more readable

Support a syntax like:

      slapos request \
        --node=computer_guid=local \
        --parameters-file=~/request.json \
        ERP5 \
        https://lab.nexedi.com/nexedi/slapos/raw/1.0.145/software/erp5/software.cfg

to request an instance with parameters in ~/request.json file.

This automatically understands the serialisation, when software uses json-in-xml
serialisation, no need to use `_` argument with the encoded json.

This allows easily serving a directory content with a server like this:

    from six.moves import SimpleHTTPServer

    class FileHTTPServer(ManagedHTTPServer):
      working_directory = profile_dir
      RequestHandler = SimpleHTTPServer.SimpleHTTPRequestHandler

This reverts commit 202a9bcd because it caused
some test failures.

   This prevents diff while commit others things

Forgot to mark the version.

It's a release which fixes changelog for badly released 1.6.15

  This changes aims minimize the unecessary accessed to master while the
  instance is going to be skipped to process or won't post informations to master.

  This in best cases reduces access from 1+n to 1 while n = number of used partitions.

  Since the status is part of the getFullComputerInformation, once it
changes state reindex the instance. This is required to prevent the
message be outdated while check for anomaly.

[collect] section was put in the middle of [networkcache] section resulting
with badly working fresh slapos-node installations using the example as a
configuration template.

This fixes commit 2da17657

See merge request nexedi/slapos.core!318

OfficeJS's user blocking third party cookie will at least stop entering their credentials many times for nothing

   If the mariadb is more recent (and consitent) them the ZODB, it isn't needed to full reindex since all
   changed objects are visible in the catalog. So it is safe enough only partially reindex.

Until now, standalone was running slapos node software/instance with --all flag
which force installing software or processing software, unlike "normal" slapos
node which no longer install on software once they are completed and only process
instances when they are requested with different parameters or when they have
failing promises.

We stop using the --all flag, to behave like a normal slapos node.

This reveal missing promises in some softwares, after this change, waitForInstance
can return faster. This reveal test failures with some softwares where the
instanciation step request other instances, but without having a promise to wait
for their requests to have been sucessfully processed; in this case waitForInstance
return too early.

We keep an "slapos node software --all" API to force reinstalling software, this
can be useful for scenarios like erp5testnode, or software release development.

We also keep an hidden "slapos node instance --all" API, so that we can keep
running tests for software releases with missing promises.

See merge request nexedi/slapos.core!312

See merge request nexedi/slapos.core!313

This test suite is for run the tests in the erp5-doc repository.
Put it in this repository is because the testpages in the
erp5-doc/erp5_slapos_tutorial need slapos-master recipe to configure the
erp5 instance.

Old slapgrid version will use the "Hosting Subscription" term during
search.
Add a new Skin Selection to rewrite the search query on the fly.

Add a new web site to not rely on the generic one, as it prevent specific customization