Use an expected list

[rgaudin]

In order to implement #114, we need the following modifications to the scraper:

• [x] build an expected_articles list/set using MW API, based on requested categories, --only and --exclude.
• [x] Skip articles scraping for those not in the list
• [x] Update rewriter to work off this new list instead of the current combination of options
• [x] Update rewriter to remove card or all missing links (should be mostly OK)
• [x] Update scrape_category_page to work with the new list
• [x] remove cards instead of link target
• [x] Update breadcrumb rewriting to work off the list