fix: correct character indexing for non-ASCII strings in markdown escaping

[CirnoV]

Fixes indexing issues when processing non-ASCII strings during markdown conversion.

• Replace value.chars().nth(*position) with value[*position..].chars().next() for proper byte-based
  indexing
• Prevents potential panics when handling multi-byte UTF-8 characters
• Update test cases