Publish compact transition schema v3

Set Apache-2.0 dataset metadata

Configure recursive Parquet data paths

docs: simplify OProofs description

docs: simplify source table

docs: clarify original source datasets

docs: document benchmark quarantine policy

docs: make dataset card general

docs: update dataset card coverage and status

Docs: explain LeanTransitionCorpus for ATP training

Document immutable streamed extraction sources

Correct globally unique example ID documentation

Publish unique example IDs and consolidate the verified local seed

Consolidate verified local seed and document its schema

Document lossless Parquet schema and clean splits

initial commit