jiayu: restore v1 split alongside v2 (v1=99, v2=74)

Restore v1 split (99 rows, full dataset)

Expose yuner v2 split in dataset card

Upload yuner v2 dataset

Update v2: 74 rows - delete templates, fix subjects (Ming hanfu, Corgi, Border Collie, etc.)

Update v2: remove seeddream-only rows (78 total)

Remove old v1 jiayu parquet

Update jiayu config to v2 split (91 rows, no-GT removed)

Add v2 split (91 rows, no-GT removed)

Upload yuner-style dataset

Delete obsolete remote folder yuner

Upload yuner-style dataset

Delete obsolete remote folder yuner

Simplify README YAML: remove dataset_info, keep configs only

update new data

Add jiayu subset config (dataset_info + configs) to README

Add jiayu subset config to README

Add jiayu subset (103 rows: train+gpt+gpt2+gpt3+seeddream)

Upload dataset

Upload dataset

initial commit