Expand parser and Unicode test coverage

test/testdata/invalid-unicode

@@ -0,0 +1,89 @@
+<EFBFBD>
+====
+-1
+unable to decode byte 0xe5 at position 0
+========
+["<22> <-- invalid UTF-8"]
+====
+-1
+unable to decode byte 0xe5 at position 2
+========
+[<5B>]
+====
+-1
+unable to decode byte 0xe5 at position 1
+========
+[a<>]
+====
+-1
+unable to decode byte 0xe5 at position 2
+========
+["\u<>"]
+====
+-1
+unable to decode byte 0xe5 at position 4
+========
+["\<5C>"]
+====
+-1
+unable to decode byte 0xe5 at position 3
+========
+[0<>]
+====
+-1
+unable to decode byte 0xe5 at position 2
+========
+[123<32>]
+====
+-1
+unable to decode byte 0xe5 at position 4
+========
+[1e<31>]
+====
+-1
+unable to decode byte 0xe5 at position 3
+========
+[1e1<65>]
+====
+-1
+unable to decode byte 0xe5 at position 4
+========
+["<22>"]
+====
+-1
+unable to decode byte 0x81 at position 2
+========
+["<22>"]
+====
+-1
+unable to decode byte 0xc1 at position 2
+========
+["<22>"]
+====
+-1
+unable to decode byte 0xfd at position 2
+========
+["<22><><EFBFBD><EFBFBD>"]
+====
+-1
+unable to decode byte 0xf4 at position 2
+========
+["<22><><EFBFBD> <-- overlong encoding"]
+====
+-1
+unable to decode byte 0xe0 at position 2
+========
+["<22><><EFBFBD><EFBFBD> <-- overlong encoding"]
+====
+-1
+unable to decode byte 0xf0 at position 2
+========
+["<22><> <-- truncated UTF-8"]
+====
+-1
+unable to decode byte 0xe0 at position 2
+========
+["<22><><EFBFBD> <-- encoded surrogate half"]
+====
+-1
+unable to decode byte 0xed at position 2