Add list of upstream datasets to each dataset doc

[nesnoj]

For each dataset in the docs, add list of preceding datasets (one level) before the metadata section.
snakemake --dag (cf. CLI docs) could be useful as it exports the DAG data. As snakemake cannot execute itself (I guess), the command must be executed beforehand. Or maybe the API can be used?