[css-fonts] Untested things from css-fonts-3

This is a first pass. Some of these might be easy to test; in particular 
a bunch of statements of the form "feature X is equivalent to these 
low-level settings A B C". I will work on a more detailed list as part 
of my action.

Untested or significantly undertested:

  * 6.2 Language-specific display
      o (no tests at all)
  * 6.3 Kerning: the font-kerning property
      o (no tests at all)
  * 6.4 Ligatures: the font-variant-ligatures property
      o (low-level equivalence tests)
  * 6.5 Subscript and superscript forms: the font-variant-position property
      o (low-level equivalence tests)
      o test for non-nesting
  * 6.6 Capitalization: the font-variant-caps property
      o (low-level equivalence tests)
      o lots of other normative statements in that section
  * 6.7 Numerical formatting: the font-variant-numeric property
      o (low-level equivalence tests)
  * 6.8 Alternates and swashes: the font-variant-alternates property
      o (low-level equivalence tests)
  * 6.9 Defining font specific alternates: the @font-feature-values rule
      o lots of other normative statements in that section
  * 6.10 East Asian text rendering: the font-variant-east-asian property
      o (low-level equivalence tests)
  * 6.12 Low-level font feature settings control: the
    font-feature-settings property
      o (all the above low-level equivalence tests also test this)
  * 6.13 Font language override: the font-language-override property
      o (test from the example in the spec?)
  * 7.1 Default features
      o enabled by default, test
  * 7.2 Feature precedence
  * 8 Object Model
      o (no tests at all) does not seem to be implemented the way the
        spec says
  * 8.1 The CSSFontFaceRule interface
  * 8.2 The CSSFontFeatureValuesRule interface

-- 
Chris Lilley
@svgeesus
Technical Director @ W3C
W3C Strategy Team, Core Web Design
W3C Architecture & Technology Team, Core Web & Media