Change the string token parsing to bump by total_len, instead of remainder[0..total_len].as_bytes().len()

[LikeLakers2]

Hi, this is my first PR here. I was looking through how taplo parses inputs, and saw the string parsing functions could be slightly optimized.

To explain my thought process for this change:

• When parsing for a string in inputs, taplo iterates over each character, adding c.len_utf8() to a variable called total_len.

  taplo/crates/taplo/src/syntax.rs

  Line 136 in 6694969

  total_len += c.len_utf8();

• char::len_utf8() returns the number of bytes that character would need in UTF-8. https://doc.rust-lang.org/1.70.0/std/primitive.char.html#method.len_utf8
• Later, when we finish parsing the string, we bump the end of the currently lexed token by remainder[0..total_len].as_bytes().len() bytes.

  taplo/crates/taplo/src/syntax.rs

  Line 144 in 6694969

  lex.bump(remainder[0..total_len].as_bytes().len());

• ...except, impl SliceIndex<str> for Range<usize> takes in a start and end point in bytes. https://doc.rust-lang.org/1.70.0/std/primitive.str.html#impl-SliceIndex%3Cstr%3E-for-Range%3Cusize%3E
• This means the length of remainder[0..total_len] in bytes will always be total_len. Thus, we can just replace remainder[0..total_len].as_bytes().len() with total_len.

I was not able to test this PR locally, as cargo test would fail due to issues compiling pprof on my Windows machine, but I did use a separate branch in my fork to run the GitHub workflows, which you can see the results of here: https://github.com/LikeLakers2/taplo/actions/runs/10974635267

If I missed anything, or if my thought process is wrong, please let me know. :) Thank you in advance for reviewing my PR!