For this release, I completely rewrote the API backend. Before, it parsed through the page HTML for entire Wikipedia articles. This was slow and inefficient, but more importantly it violated the Wikimedia bot policy (https://meta.wikimedia.org/wiki/Bot_policy#Unacceptable_usage). Now, this project uses the official MediaWiki API, which has lots of advantages (see latest commit message)