- Update the debian/dirs files.
- Change the version and add the version in the name or field where required.

templates/*: Improve the templates system: It is no longer needed to modify the templates to build another SR.
templates/Makefile.in: Improve the target management.
build-scripts/*: Add new variables and improve the genericization.

build-scripts/*: Simplify the scripts and add a variable for the path of the current tarball.
build-scripts/obs.sh: Delete .install.cfg in the directory where builtout is run.

- Fix a bug:
	Transform templates/Makefile in an actual template Makefile.in leaving less room for path inconsistencies.
- Fix a bug. Remove tmp-networkcached-related commands.

templates/buildout*.in:
- Fix a bug:
	Add the extends-cache and download-cache variables.

build-scripts/template_stage.sh:
	Add support for the new templates/Makefile.in

build-scripts/obs.sh: Fix a bug. Save the path of the local BUILD_DIRECTORY so that it can be replaced in the relevant files by the right path in OBS' VM.
dependencies.txt: Add a dependency.

- README.md: Improve the advice for installing osc
- Add dependencies.txt which lists the dependencies required by the project.

Previously: lunzip.
TODO: Make this process more intuitive and/or document it.

- Remove the top-level Makefile of the archive.
- Rename directories.
- Send directly tarballs/$SOFTWARE.
- Build from tarballs/$SOFTWARE/Makefile.
- Rename variables for clarity.
- Add obs.sh and clean_obs.sh.

- remove template/networkcached.cfg.in
- remove the networkcached support from build_tree.sh and clean_build_tree.sh
- remove the networkcached support from template_stage.sh bootstrap_buildout.sh
- remove cache_creation.sh and clean_cache_creation.sh

The scripts do everything until the first run of buildout (actual run, not just bootstraping).

Previously it only worked with Debian 10. This commit reflects the
removal of the apt package 'python-pip' in Debian 11 and the change
of the i2c bus number.

There were still two issues which are solved with this commit:

1. Previously the firewall prevented the installation of Fluentd
   (because no connection via http/https was possible).

2. Previously the sensor couldn't send data to the IoT gateway (
   the opened port wasn't sufficient).

When importing the slapos playbook after running re6stnet-install role,
the user defined computer_name will be ignored and will be overridden
by 'set_facts', line 16 in re6stnet-install/tasks/main.yml. Therefore
the slapos install role is directly called (instead of importing the
playbook) which seems to solve the issue.

(due to activated firewall)

(wrong indentation)

(wasn't possible because of too restrictive ipv6 firewall)

Add restrictive firewall settings to avoid any security risk.

Bug appeared due to wrong pip version. Therefore the version has been
set explicitly.

Now the Python script name is consistent with the file name in the
Fluentd config file as described in the learning track.

Should be installed via rapid.space panel.

Previously it failed because setuptools wasn't installed.

Replaced 'computer_name' with 'computer_id'. Previously I confused both.
Now the user has to enter the respective computer reference.

It won't run in an issue anymore if the file already exists.