Resplit dataset with 8:2 train/test ratio while maintaining subject_code distribution

Update README.md

Update dataset_infos.json

Add Classical Poetry Subject Clustering Benchmark - 题材 Clustering Benchmark

Delete dataset_infos.json

Delete README.md

Delete test-00000-of-00001.parquet

Add Classical Poetry Subject Clustering Benchmark - 题材 Clustering Benchmark

initial commit