[BUILD] Generate synthetic test data: text files and pattern sets

[hubertwojcik]

Goal

Create scripts that generate reproducible synthetic text files and pattern sets of varying sizes for all benchmarking experiments.

Tasks

• Write scripts/gen_text.py: generate random text files of configurable size (10 MB / 100 MB / 1 GB)
• Write scripts/gen_patterns.py: generate random pattern sets of configurable count (10 / 100 / 1000 patterns) and length
• Seed RNG for reproducibility; document seed in benchmarks/README.md
• Add a make data target that generates the standard benchmark dataset

Acceptance Criteria

• Scripts generate files of the exact requested size
• Pattern files contain no duplicates
• Generated data is excluded from git via .gitignore; generation is fully reproducible from seed