The previous logic parses URLs with a regex first and then using
the markdown library. It then uses urlopen().read() to validate
links.

We use now the markdown library only to extract the list of links,
and then urlparse to deconstruct, analyse, adapt and reconstract
the link. We do not attempt to fetch links anymore, which means
that external links are not guaranteed to be working.

Absolute URLs are not changed (they may be external)
Fragments are relative to the page and do not need changes
Path only links should point to a file synced to the website
but sometimes the file may be missing (if it's not in the sync
configuration), so we follow this approach:
- prefix with base_path and check for the file locally
- if not found, prefix with base_url instead

Note that urlparse treats URLs without scheme like path only
URLs, so 'github.com' will be rewritten to base_url/github.com
Signed-off-by: NAndrea Frittoli <andrea.frittoli@gmail.com>

As a developer, it is difficult to make changes to a script and ensure that nothing is broken without tests. Test are added in this patch with 80% code coverage. The script has also been refactored to increase readability. This refactor is not a pure refactor. The sync did not work properly. There is a list of bugs that reported the script generating broken links, #160, #126, #158, #133, #134. This refactor also includes a patch that fixes these bugs. The patch redirects a user to github if there is a relative link to a file that does not exist. This is important because some links will reference a snippet of code, test, or example. These artifacts are not docs, they are not rendered properly on the site. It is better to redirect the user to the intended artifact if the file is not synced. This prevents random 404s, and preserves the original behavior of the sync script. There are other approaches to fix this bug. This includes, manually editing links, downloading the files that the broken link are referencing in the sync config file, or removing the broken links. This is the best approach because broken links are automatically fixed and do not require manual intervention. This approach causes a side effect of not requiring the config file to be up-to date because the script will redirect the user to github. We want to encourage frequent sync. So eventually, this script should become tekton-ified (tekton task) and placed into the tekton catalog repo. Sync needs to become a tekon task in order for every repo to sync docs on push event.

This achieves a few things:

1. Creates Dockerfile for sync tool, so that it can be invoked from
   Tekton tasks.
2. Makes sync config directory configurable by flag, so that the tool
   can be ran from any directory. This defaults to sync/config (same as
   before).

Symlinks are retainined in the root directory for requirements.txt and
runtime.txt to satisfy Netlify Build requirements:
https://docs.netlify.com/configure-builds/manage-dependencies/#python

Issue: #34