`float` member issue with `clang++-15` and `g++-12`

[t-mat]

`float` member issue with `clang++-15` and `g++-12`

[stephenberry]

@t-mat
This was actually simpler than I thought it might be, so I was able to fix it right away in . Thanks again for reporting this issue, it's been added as a unit test.

One note of warning is that because you append a null character to your serialized buffer, when you copy to your output string and use ser.size(), this will include your null character in your output string, you'll get two null characters at the end of your string and your string's size will include the first null character. It's usually better to simply use std::string as your buffers to avoid these kinds of bugs and not need to append your own null characters.

[stephenberry]

This has now been fully fixed as of #1011. The floating point conversion code hadn't supported std::vector<std::byte> so the float was being casted to an int. Glaze now blocks these implicit conversions and correctly supports std::vector<std::byte>

[stephenberry]

Thanks for sharing this bug. I should be able to fix it tomorrow.