[feature] Unicode property escapes not supported (\p{L}, \p{N}, etc.)

[jbachorik]

Summary

Unicode property escapes (\p{L}, \p{N}, \P{L}, etc.) are not supported. Patterns using them are currently rejected or silently mis-matched.

Examples

• \p{L} — any Unicode letter
• \p{N} — any Unicode number
• \p{Lu} — uppercase letter
• \P{L} — negated: any non-letter

Impact

• 66 PCRE tests are currently filtered out entirely because they use unsupported features including Unicode properties
• Common in real-world patterns for internationalized text

Implementation Notes

• Difficulty: Medium-High (large Unicode category tables required)
• Files: RegexParser.java (parse \p{...}), new UnicodePropertyCharClass AST node, ThompsonBuilder.java, charset integration
• Unicode data can be derived from JDK's Character class to avoid external dependencies