Check why UTF-8 decoding does not fail on illegal encodings of Surrogate codes

[cowtowncoder]

See https://github.com/FasterXML/jackson-dataformats-binary/pull/65 for background: it looks like UTF-8 json parser is not checking for case of 3-byte character being surrogate: I think it should. (although there is still the whole issue of why JSON escapes may be used in such a way...).