XML parser incorrectly accepts document without top-level element

Imported from http://bugzilla.roxen.com/bugzilla/show_bug.cgi?id=3932

Reported by Jonas Wallden jonasw@roxen.com

Both of these examples should signal parse errors:

  ceylon:server $ ~/pike/7.4/bin/pike
  Pike v7.4 release 301 running Hilfe v3.5 (Incremental Pike Frontend)
  > function cb = lambda(mixed ...foo) { werror("%O\n", foo); };
  > Parser.XML.Simple()->parse("", cb);
  (1) Result: ({ })
  > Parser.XML.Simple()->parse("<?xml version='1.0'?>", cb);
  ({ /* 5 elements */
      "<?xml",
      0,
      ([ /* 1 element */
        "version":"1.0"
      ]),
      0,
      ([ /* 1 element */
        "location":21
      ])
  })
  (2) Result: ({ })

Relevant productions from the XML specification:

[1] document ::= prolog element Misc* [39] element ::= EmptyElemTag | STag content ETag [40] STag ::= '<' Name (S Attribute)* S? '>' [44] EmptyElemTag ::= '<' Name (S Attribute)* S? '/>'