This removes files that were providing pkglists redirection for updater
versions in 4.0-alpha* releases.

First officially forward compatible version is 4.0-beta1 and that does
not have any pkglists in medkit while still expecting these lists in old
location. The advantage is that in medkit those lists are not enabled so
update is possible and nobody should still run 4.0-beta1 now so we can
drop this without a problem.

This allows access to file browser in Luci but this is very limited
access. There seems to be a problem with displaying files and there is
no way to modify files at all so this is kind of useless. This is also
not directly accessible trough Luci but you have to know specific URI to
access it. This makes it probably just fully unknown to users so let's
just drop it.

This does not work and contains non-existing packages. Let's drop this.

These comments are effectively just repeating what code says. It should
be pretty much clear from code and file name what specific include
contains so no comment is needed.

dnl is left in place to eat up new line character after include
directive.

This reverts to previous approach for Luci localization packages. It
introduces back problems that language packages are installed only to
apps introduced by updater lists but there is no way around it. This
implementation was problematic as it kept installed packages in place
because of language packages depending on package that was suppose to be
removed. This meant that such package was hold in place by language
package that was in turn hold in place by information that parent
package is installed. It created cycle that updater was not able to
break (only opkg). This fixes this issue.

This code was causing problems with packages not being removed because
they were held on place by localization package. This reverts it to a
approach where we specify language packages only if we add request for
package it self.

These options are required as there seems to be no default set for those
options.

stat and diff are base utilitites that are required by even core files
such as /etc/rc.common so they should be part of base-minimal
installation.

This just sorts them and comments them.

It also drops Moxtet options as those were probably added by mistake.

There is a problem with switching away from kmod-ath10k-ct on Mox as
kmod-ath10k provide ct variant but not the other way around. This fixes
that and allows users to switch no non-ct drivers while ensuring that at
least one driver is installed and on Mox the ct one is default one.

btrfs-progs now use preinit instead of this init script so we no longer
have to activate it.

This drops patch that modified default kernel configuration for Turris
1.x and instead moves all relevant options to Turris 1.x specific
config.

Upstream renamed it to config.buildinfo.

They also introduced new file feeds.buildinfo. We could use that instead
of our git-hash generated file.

This moves all options from Omnia support patch to common configs. They
are all part of kernel support for packages so it should be enabled on
all our routers. This ensures that. It also means that there is one less
patch to rebase and maintain.

This also moves some stuff around in common kernel configuration.

This patch is not valid because it patches 4.14 kernel but mvebu is
switched to 4.19 but for now let's have it as it is before compilation
of tools for aarch64 is fixed.

This package was disabled in 9098d37c
without explanation. Let's try to reebable it to see if it works or what
it breaks.

Upstream changed names of variables in /etc/os-release to correspond to
correct project name. It is no longer Lede but once again OpenWRT.

Configuration override options was misleading. In reality those are
options you most probably don't want to mess too much with so let's call
them rather advanced options.

This also moves PUBLISH_BRANCH in example configuration file to
section: don't mess with. This is because now we provide it from
defaults and user just should not mess with it unless he know what he is
doing.

OpenWRT commit to be used is specified in feeds.conf file with rest of
feeds now so it makes not sense to overwrite it. This is just confusing
without any real gain in here. You can always modify feeds.conf file to
achieve the same.