Collect statistical data while parsing

I recently added an `analyzer.rb`, which processes existing errata-files and collects statistical data.

This was also used in `check_latest_errata.rb`, which is supposed to provide a nagios-check.
However, the analysation of the files takes too long for nagios, so it would be better to collect the statistical data during errata creation.