Add scraped page archive

[ondenman]

What does this do?

Uses scraped-page-archive to archive all pages scraped.

Why is this needed?

There's an election coming up (01/12/2016), and it's likely that the data on the official site will disappear, meaning any data we're not already picking up will be lost. Archiving it now gives us the chance to go back and re-scrape later even if it disappears.

Relevant Issue(s):

everypolitician/everypolitician-data#20544

Checklists:

Scraper change:

• scraper is on Morph.io under the "everypolitician-scrapers" group — no, it's still at https://morph.io/tmtmtmtm/gabon-deputes, but there's little point in moving it this close to the election.
• scraper's GitHub "Website" link points at morph.io page
• scraper is set to auto-run — not yet: @tmtmtmtm This needs to be set on Morph.
• scraper is configured for archiving — that's what this PR is doing!

Adding Archiving:

• scraper uses scraped-page-archive gem directly or via a suitable strategy — uses it directly
• MORPH_SCRAPER_CACHE_GITHUB_REPO_URL is configured — not yet. @tmtmtmtm this needs configured on morph
• pages are being archived in new branch of correct scraper repo (yay!)

Gemfile change:

• all links are secure
• links to Github use github: protocol, not simply git:
• formatting is consistent with our normal Rubocop setup (See commit message.)

[tmtmtmtm]

Doesn't include https://github.com/everypolitician/everypolitician/wiki/Scraper-Change-checklist

[tmtmtmtm]

@ondenman the rubocop tidying only needs to be for the Gemfile, not the scraper.rb. As the autofix actually changes the logic in the scraper (by hoisting the guard clause), I'm a little hesitant to include this here. I'd suggest dropping 737f9ac, and rerunning separately against only the Gemfile.

(NB you don't necessarily need to actually include Rubocop+config in these sort of changes, but there's no harm in doing so, so now that it's already here it's OK to keep it)

[tmtmtmtm]

The PR description here is a bit misleading — this isn't really archiving any pages, only a PDF file which presumably is never going to change. As such this change probably isn't that useful (other than a single run to archive that PDF once).

However, I suspect we should also archive (though not process) http://www.assemblee-nationale.ga/34-deputes/168-bureaux-des-commissions/ and http://www.assemblee-nationale.ga/34-deputes/153-les-femmes-deputes/

I've added an extra commit to pick up those two pages.

[tmtmtmtm]

I've configured morph, though I haven't set it to run every day, I doubt it's actually going to change again between now and the election.