Description

This pull-request is a rewrite of the comment-header at the start of languages.yml, which currently serves as the file's only canonical documentation. Aside from being ordered arbitrarily, it also contains some falsehoods sitting in plain sight:

1. type - Either data, programming, markup, prose, or nil

   As we all know, nil isn't a valid language type. This field is mandatory.

2. color - CSS hex color to represent the language. Only used if type is "programming" or "markup".

   This hasn't been true for a while. These colours are used to identify the languages of search results, and are also visible if linguist-detectable is used to include data and prose languages in language breakdown graphs.

Other improvements include:

1. Eliminating references to Ruby jargon (e.g., name.downcase may be obvious to a Ruby programmer, but less so to a novice coder who's only ever seen toLowerCase() used, for example).
2. Ordering fields alphabetically¹, and grouping them by their requirement status.
3. Referring to field values in more generic terms, rather than the RubyDoc convention of capitalising class-names that aren't proper nouns (e.g., a String name, an Array of, etc).
4. Clarifying what certain fields mean for those less familiar with the sacred ways of Unix (a Windows user might misinterpret the, uh, interpreters field if they've only ever worked in an IDE). This holds especially true for anybody whose English may not be perfect.

Please let me know if anything sticks out as strangely-worded or whatever. I spent so much time juggling words and resmithing paragraphs, I developed tunnel vision. It may need a second pair of eyes that aren't glazing over like mine are right now.