HTTP parser: relaxed checking of fields values.

Allowing characters up to 0xFF doesn't conflict with RFC 7230.
Particularly, this make it possible to pass unencoded UTF-8 data
through HTTP headers, which can be useful.

src/nxt_http_parse.c

@@ -699,8 +699,7 @@ nxt_http_lookup_field_end(u_char *p, u_char *end)
 
 #define nxt_field_end_test_char(ch)                                           \
                                                                               \
-        /* Values below 0x20 become more than 0xDF. */                        \
-        if (nxt_slow_path((u_char) ((ch) - 0x20) > 0x5E)) {                   \
+        if (nxt_slow_path((ch) < 0x20)) {                                     \
             return &(ch);                                                     \
         }
 

src/test/nxt_http_parse_test.c

@@ -301,7 +301,7 @@ static nxt_http_parse_test_case_t  nxt_http_test_cases[] = {
     {
         nxt_string("GET / HTTP/1.1\r\n"
                    "Host: пример.испытание\r\n\r\n"),
-        NXT_HTTP_PARSE_INVALID,
+        NXT_DONE,
         NULL, { NULL }
     },
     {