Add initial content for explanation/merges and syncs

[dviererbe]

This PR adds the initial content for the explanation of syncs and merges from Debian unstable.

[dviererbe]

I looked through the CI errors:

• synchronization is marked as a spelling error. Most likely, because it wants it to be written as synchronisation 🙄
• The refs DebianImportFreeze, FeatureFreeze, FinalRelease, etc. are not defined, because the follow up PR Add initial content for explanantion/development process #54 does it. This also causes spell checking errors.
• Link check times out for https://merges.ubuntu.com/*. The long loading time for Merge-o-Matic is expected behavior. Maybe we should add a linkcheck ignore. Not ideal, but acceptable for me. We could also look into massively increasing the timeout time. Don't know what the value should be so that it is not flaky.
• The spelling checker also does not like the "Matic" of "Merge-o-Matic" and its acronym "MoM". This should be easily fixable.
• The manpages.ubuntu.com linkckecks are marked as 404 broken, but i can not figure out why. They work for me in a browser.

[s-makin]

* Link check times out for `https://merges.ubuntu.com/*`. The long loading time for Merge-o-Matic is expected behavior. Maybe we should add a linkcheck ignore. Not ideal, but acceptable for me. We could also look into massively increasing the timeout time. Don't know what the value should be so that it is not flaky.

I would agree with adding it to the linkcheck ignore. I think that would be better than increasing the timeout time, otherwise people working on PRs or making contributions will end up having frustrating experiences with the tests.

* The `manpages.ubuntu.com` linkckecks are marked as 404 broken, but i can not figure out why. They work for me in a browser.

I was having the same issue with manpages.ubuntu.com links in Pro docs. I couldn't see any legitimate reason for it, and the links were working when I clicked on them. I'd be a bit more hesitant about adding that to linkcheck ignore though, since there might be genuinely broken pages among those, which we will want to at least be prompted to check.

[dviererbe]

I'd be a bit more hesitant about adding [manpages.ubuntu.com] to linkcheck ignore though, since there might be genuinely broken pages among those, which we will want to at least be prompted to check.

Fully agree, but I also do not know how to proceed here. Manually checking links every time can't be the solution :/

[dviererbe]

I'd be a bit more hesitant about adding [manpages.ubuntu.com] to linkcheck ignore though, since there might be genuinely broken pages among those, which we will want to at least be prompted to check.

Fully agree, but I also do not know how to proceed here. Manually checking links every time can't be the solution :/

suggestion: I could write a script that greps all :manpage:`value` and check it with curl and tell linkcheck to ignore manpage.ubuntu.com.

[dviererbe]

I through together this script. It even has the benefit of de-duplicating manpage checks.

#!/usr/bin/env bash

manpages=$(grep \
    --basic-regexp ':manpage:`.\+`' \
    --only-matching \
    --recursive --include '*.rst' docs \
    | cut --delimiter '`' --fields 2 \
    | sort --unique)
    
pass=true

for manpage in $manpages
do
    # assuming the pattern "PAGE(SECTION)"
    page="${manpage%%(*}"
    section="${manpage#*(}"
    section="${section%)*}"

    url="https://manpages.ubuntu.com/manpages/en/man${section}/${page}.${section}.html"

    if curl --silent --show-error $url > /dev/null
    then
        echo "INFO: Tested $manpage successfully (URL='$url')."
    else
        pass=false
        echo "ERROR: Failed to load manpage $manpage (URL='$url')."
    fi
done

if $pass
then
    echo "PASS: All manpages were accessible."
else
    echo "FAIL: One or more manpage(s) did not load properly."
    exit 1
fi