[css-syntax] Drop the explicit mention of the @charset rule, since such a rule doesn't exist. Fixes #3464.

Author: tabatkins

css-syntax-3/Overview.bs

@@ -2025,11 +2025,6 @@ Parse a stylesheet</h4>
 			<a>Consume a list of rules</a> from the stream of tokens, with the top-level flag set.
 			Let the return value be <var>rules</var>.
 
-		<li>
-			If the first rule in <var>rules</var> is an <a>at-rule</a> with a name
-			that is an <a>ASCII case-insensitive</a> match for "charset",
-			remove it from <var>rules</var>.
-
 		<li>
 			Assign <var>rules</var> to the stylesheet's value.
 
@@ -3228,13 +3223,6 @@ The ''@charset'' Rule</h3>
 	any occurrences of an ''@charset'' rule must be treated as an unrecognized rule,
 	and thus dropped as invalid when the stylesheet is grammar-checked.
 
-	Note: The algorithm to <a>parse a stylesheet</a> explicitly drops the first ''@charset'' rule from the document,
-	before the stylesheet is grammar-checked,
-	so valid rules that must appear first in the stylesheet,
-	such as ''@import'',
-	can still be preceded by an (invalid) ''@charset'' rule
-	without making themselves invalid.
-
 	Note: In CSS 2.1, ''@charset'' was a valid rule.
 	Some legacy specs may still refer to a ''@charset'' rule,
 	and explicitly talk about its presence in the stylesheet.
@@ -3447,18 +3435,23 @@ Changes from the 20 February 2014 Candidate Recommendation</h3>
 	* Fix several bugs in the tokenization and parsing algorithms.
 
 	* Change the definition of ident-like tokens to allow "--" to start an ident.
+		As part of this, rearrange the ordering of the clauses in the "-" step of <a>consume a token</a>
+		so that <<CDC-token>>s are recognized as such instead of becoming a ''--'' <<ident-token>>.
 
 	* Don't serialize the digit in an <<an+b>> when A is 1 or -1.
 
 	* Define all tokens to have a <a>representation</a>.
 
 	* Fixed minor bug in <a>check if two code points are a valid escape</a>--
 		a <code>\</code> followed by an EOF is now correctly reported as <em>not</em> a valid escape.
+		A final <code>\</code> in a stylesheet now just emits itself as a <<delim-token>>.
 
-	* If the very first rule in the sheet is a ''@charset'', dropped it during parsing.
+	* @charset is no longer a valid CSS rule (there's just an encoding declaration that <em>looks</em> like a rule named @charset)
 
 	* Trimmed whitespace from the beginning/ending of a declaration's value during parsing.
 
+	* Removed the Selectors-specific tokens, per WG resolution.
+
 	The following editorial changes were made:
 
 	* The "Consume a string token" algorithm was changed to allow calling it without specifying an explicit ending token,
@@ -3471,8 +3464,6 @@ Changes from the 20 February 2014 Candidate Recommendation</h3>
 
 	* Added the <<declaration-value>> and <<any-value>> productions.
 
-	* Removed the Selectors-specific tokens, per WG resolution.
-
 	* Removed "code point" and "surrogate code point" in favor of the identical definitions in the Infra Standard.
 
 	* Clarified on every range that they are inclusive.