reject out-of-range code points in UTF-32 wide-string input

[Angadi56]

wide_string_input_helper<BaseInputAdapter, 4>::fill_buffer converts UTF-32 input into the UTF-8 the lexer reads, and any code point above U+10FFFF falls into an "unknown character" branch that passes the code unit straight through. That pass-through narrows a 32-bit unit to int, so the single unit 0xFFFFFFFF arrives as -1, which is exactly std::char_traits<char>::eof(). The lexer reads that as end of input, so the parse stops there and the strict trailing check in parser::parse still succeeds: json::parse(std::u32string(U"[1]") + char32_t(0xFFFFFFFF) + U"anything") returns [1] and json::accept returns true, with every unit after the sentinel silently discarded. I found it while lining this branch up against the sibling UTF-16 helper, whose pass-through is safe only because its units cannot exceed 0xFFFF. The neighboring out-of-range values already behave correctly, since 0x110000 and a lone surrogate both stay positive after the cast and raise parse_error.101, so only the one value that collides with the sentinel is mis-decoded. The fix has that branch emit a byte that is never valid UTF-8 instead of a narrowed unit, which is the same collision input_stream_adapter::get_character already documents guarding against, and it leaves the two in-range casts in the file alone because they are bounded to 0..0x7F. Valid input is unchanged and the regression test goes in the existing invalid std::u32string section.

• The changes are described in detail, both the what and why.
• If applicable, an existing issue is referenced.
• The Code coverage remained at 100%. A test case for every new line of code.
• If applicable, the documentation is updated.
• The source code is amalgamated by running make amalgamate.

[nlohmann]

Good catch! The change looks good, but please fix the self-cast warning from GCC.

[nlohmann]

Clang was updated and the latest develop branch fixes a CI issue. Please rebase to the latest develop branch.