Add self-updating ability

If your workflow uses GitHub releases, with version 1.9 of Alfred-Workflow it will be able to update itself.

See the documentation for information on setting up and using the self-updating feature.

Thanks to @fniephaus

Reorganise documentation

Move documentation out of source code and into the docs.

Split docs into several smaller files.

Change theme to Read the Docs theme.

Mirror documentation on Read the Docs.

Clarify behaviour of Workflow.filter()

Workflow.filter() now raises a ValueError if query is empty or a blank string.

Improve and clarify documentation.

Documentation is now also available on Read the Docs (beta).

Fix filter sort regression

Fix #34

Work from within non-ASCII directories

Fix bug #29 where non-ASCII characters in the path to the workflow's directory would cause Workflow to fail.

Add support for Large Type and Copy Text

Add largetext and copytext arguments to Workflow.add_item().

If Script Filter results have a specified copytext, this will be copied to the clipboard (instead of arg) when CMD+C is pressed on the item.

If results have a specified largetext, this will be shown in Alfred's large text pop-up (instead of the query) when CMD+L is pressed on the item.

Allow workflow.py to work from outside the workflow

This will enable Alfred-Workflow to be installed via pip and the bundler.

Improve encoding detection of HTTP resources

• Better parsing of HTTP headers
• Better sniffing of HTML/XML document headers
• Better fallback default encodings for common mimetypes

Improve docs; make Settings.save() method public

Add improved documentation regarding settings, stored data, background processes

Add data storage methods and custom serializers

• Add new Workflow.store_data(name, data) and Workflow.stored_data(name) methods.
• Add SerializerManager to enable custom serialization formats for stored and cached data.

Implements #17