Add another degenerate case for r-g to handle circles, as they're currently discontinuous in a fixable way.

Author: tabatkins

css3-images/Overview.html

@@ -1212,6 +1212,14 @@ <h4 class=no-toc id=degenerate-radials><span class=secno>4.2.3. </span>
    degenerate cases, the gradient must be be rendered as follows:
 
   <dl>
+   <dt>If the <a href="#ending-shape"><i>ending shape</i></a> is a circle
+    with zero width or height:
+
+   <dd>Render as if the <a href="#ending-shape"><i>ending shape</i></a> was a
+    circle whose radius was an arbitrary very small number greater than zero.
+    <span class=note>This will make the gradient continue to look like a
+    circle.</span>
+
    <dt>If the <a href="#ending-shape"><i>ending shape</i></a> has zero width
     (regardless of the height):
 

css3-images/Overview.src.html

@@ -747,6 +747,9 @@ <h4 class="no-toc" id="degenerate-radials">
 	be rendered as follows:</p>
 
 	<dl>
+		<dt>If the <i>ending shape</i> is a circle with zero width or height:</dt>
+		<dd>Render as if the <i>ending shape</i> was a circle whose radius was an arbitrary very small number greater than zero.  <span class='note'>This will make the gradient continue to look like a circle.</span></dd>
+
 		<dt>If the <i>ending shape</i> has zero width (regardless of the height):</dt>
 		<dd>Render as if the <i>ending shape</i> was an ellipse whose height was an arbitrary very large number and whose width was an arbitrary very small number greater than zero.  <span class='note'>This will make the gradient look similar to a horizontal linear gradient that is mirrored across the center of the ellipse.  It also means that all color-stop positions specified with a percentage resolve to ''0px''.</span></dd>