Do not hardcode chunk size in send_sample_id_and_label

[MaxiBoether]

Currently, we hard code a chunk size of 1000000 in the storage send_sample_id_and_label function. We need to chunk the number of files since potentially the number of files we get as input is huge. However, we do not really have a good parameter in the configuration (I think) to reflect the number here. Should we add one? Can we use some other parameter like sample_dbinsertion_batchsize?