hubData 2.2.2

• Fixed bug where connect_hub() returned an empty connection (with only a "No files of file format found" warning) for cloud (S3) hubs whose admin.json declares a non-parquet submission format such as csv (#148). The hubverse cloud sync always writes model-output/ to parquet on S3, regardless of the declared submission format, so connect_hub() now also accepts parquet on S3 when file_format is not supplied. Local hubs are unaffected, and hubs that genuinely store the declared format on S3 continue to read exactly as before.