Hoard 2.0.0

Hoard and Hoard Downloader are now one app, and nothing needs a command prompt or a settings file.

One app

• Library and Downloads are two views of the same app, with one set of sign-ins, tags and settings.
• Download from the page: Download on each store, Download everything new, or Download a copy
  on a single item. Progress shows as it goes, and Stop pauses safely; it carries on next time.
• Items you already have say On disk, with a link straight to them in Downloads.
• A Settings panel: the downloads folder, which stores to include (and Booth gifts, archived Gumroad
  purchases), saving images for offline use, and the browser used for store sign-ins.
• On Windows, store sign-ins use Microsoft Edge, which every PC has and Windows Update keeps patched, so
  setup no longer downloads a separate 150 MB browser. Hoard's own browser is used only if Edge is missing.

Where things live

• Settings, the library list and cached images moved into Hoard's app-data folder, next to your sign-ins
  and tags. Downloads go to a Hoard folder in Documents unless you choose another in Settings.
• Coming from 1.x: sign-ins and tags carry over by themselves. Hoard.bat migrate <old folder> brings
  over the library list and downloads folder.
• One release zip, Hoard-<version>.zip, replaces the three. Start it with Hoard.bat (or ./run.sh).

Command line

• Every command from both tools, as Hoard.bat <command> (or ./run.sh <command>): sync, verify,
  login, refresh and the rest. See docs/COMMAND-LINE.md. browse is gone: that's the Downloads view.

Under the hood

• The code is one package, hoard/, with each piece in exactly one place. The security code used to be
  copied into two or three files, kept identical by tests; now there's one copy, and a test checks it.
• Fixed: recognising which store a saved library page came from only worked for pages saved from the exact
  library address, because a second definition silently replaced the first.