Parser for ed.gov

[nightsh]

Create a HTML parser that integrates into the edgov scraper and replaces the existing dummy parser script.

Tasks:

• [x] identify the possible page structures in the target site
• [x] write one or multiple parsers that cover the most frequent cases
• [x] test if it runs well within the pipeline

Acceptance criteria:

• [x] we have a parser specific to this site that replaces the initial dummy parser
• [x] parser creates Dataset instances, each having at least one Resource
• [x] parser uses scrapy pipelines to dump datasets with resources