Fast regex search: indexing text for agent tools

Introduced a new approach to accelerate regex search for agent tools by indexing text using trigrams and inverted indexes. This builds upon existing secure codebase indexing capabilities. The system decomposes regular expressions into trigrams and uses an inverted index to find potential matches, significantly speeding up searches in large codebases compared to traditional tools like ripgrep. It also discusses suffix arrays as an alternative indexing method.