text-freetype2: Fix file text truncation at embedded null bytes

[4RH1T3CT0R7]

Strip embedded null bytes from the UTF-8 file buffer before string conversion in the Text (FreeType 2) source plugin.

A new remove_utf8_nulls() helper function is added (mirroring the existing remove_cr() pattern) and called in both load_text_from_file() and read_from_end()
after fread() and before strlen() is used.

Motivation and Context

When "Read from file" is enabled, files containing embedded null bytes (e.g. from certain text editors or encoding artifacts) are truncated at the first \0
because strlen() treats it as end-of-string. This causes only the first line or partial content to be displayed.

Since \0 never appears as a continuation byte in valid UTF-8 sequences, stripping null bytes is safe and requires no changes to downstream rendering code.

Fixes #7595

How Has This Been Tested?

• Created a test file with embedded null bytes between lines ("Line1\x00\x00Line2\nLine3") — all text now displays correctly
• Tested with a normal multi-line UTF-8 file (no null bytes) — no regression, displays as before
• Tested the "Chat log mode" (read_from_end path) with both null-byte and clean files
• Tested with an empty file — no crash, displays empty as expected

Types of changes

• Bug fix (non-breaking change which fixes an issue)

Checklist:

• My code has been run through https://github.com/obsproject/obs-studio/blob/master/.clang-format.
• I have read the https://github.com/obsproject/obs-studio/blob/master/CONTRIBUTING.rst.
• My code is not on the master branch.
• The code has been tested.
• All commit messages are properly formatted and commits squashed where appropriate.
• I have included updates to all appropriate documentation.