UTF-8 characters get garbled after formatting

[yowcow]

Describe the bug
UTF-8 characters get garbled after formatting.

To Reproduce
Have a piece of unformatted code and/or comment that includes multibyte UTF-8 characters, then trigger code formatting.

E.g.

                    %% コメント
    _ = <<
          "こんにちは"/utf8
        >>,

Expected behavior
Get a clean piece of code.

E.g.

    %% コメント
    _ = <<"こんにちは"/utf8>>,

Actual behavior
Get a garbled piece of code.

E.g.

    %% ã<82>³ã<83>¡ã<83>³ã<83><88>
    _ = <<"ã<81><93>ã<82><93>ã<81>«ã<81>¡ã<81>¯"/utf8>>,

Context

• erlang_ls version (tag/sha): Version: 0.44.1 (ref: b151f9d9a4a1dde1118ae2822226500e2fa69125)
• Editor used: neovim v0.8.0
• LSP client used: Nvim LSP client

---

This patch resolves the problem: main...yowcow:erlang_ls:fix-garbled-utf8-characters

[plux]

Fixed by #1496