Added functionality to load and process nested `.code-tokenizer-md-ignore` files for excluding patterns during markdown generation. Introduced new dependencies and ensured initialization before file processing operations.
