Avoid using the ambiguous \h shorthand character

Depending on the regular expression engine used, \h does not always
mean the same. With a PCRE engine, it matches white spaces, whereas,
with a Oniguruma engine, it matches hexademical digit characters.
Atom uses an Oniguruma engine, but github.com relies on a PCRE
engine.

Author: pchaigno

grammars/javascript.cson

@@ -1524,7 +1524,7 @@
             'name': 'punctuation.definition.unicode-escape.end.bracket.curly.js'
       }
       {
-        'match': '\\\\(x\\h{2}|[0-2][0-7]{0,2}|3[0-6][0-7]?|37[0-7]?|[4-7][0-7]?|.)'
+        'match': '\\\\(x[0-9A-Fa-f]{2}|[0-2][0-7]{0,2}|3[0-6][0-7]?|37[0-7]?|[4-7][0-7]?|.)'
         'name': 'constant.character.escape.js'
       }
     ]

grammars/regular expressions (javascript).cson

@@ -14,7 +14,7 @@
         'name': 'constant.character.character-class.regexp'
       }
       {
-        'match': '\\\\([0-7]{3}|x\\h\\h|u\\h\\h\\h\\h)'
+        'match': '\\\\([0-7]{3}|x[0-9A-Fa-f]{2}|u[0-9A-Fa-f]{4})'
         'name': 'constant.character.numeric.regexp'
       }
       {
@@ -107,7 +107,7 @@
                 'name': 'constant.character.control.regexp'
               '6':
                 'name': 'constant.character.escape.backslash.regexp'
-            'match': '(?:.|(\\\\(?:[0-7]{3}|x\\h\\h|u\\h\\h\\h\\h))|(\\\\c[A-Z])|(\\\\.))\\-(?:[^\\]\\\\]|(\\\\(?:[0-7]{3}|x\\h\\h|u\\h\\h\\h\\h))|(\\\\c[A-Z])|(\\\\.))'
+            'match': '(?:.|(\\\\(?:[0-7]{3}|x[0-9A-Fa-f]{2}|u[0-9A-Fa-f]{4}))|(\\\\c[A-Z])|(\\\\.))\\-(?:[^\\]\\\\]|(\\\\(?:[0-7]{3}|x[0-9A-Fa-f]{2}|u[0-9A-Fa-f]{4}))|(\\\\c[A-Z])|(\\\\.))'
             'name': 'constant.other.character-class.range.regexp'
           }
           {