v0.5.3

Hotfix

• Fix UTF-8 panic: Prevent panic when truncating multi-byte UTF-8 content (e.g., Chinese/CJK characters from Baidu results). Replaced byte-level slicing with safe truncate_str() using char_indices().
• Added 7 tests for UTF-8 truncation edge cases (Chinese, emoji, mixed CJK)

Full Changelog: v0.5.2...v0.5.3