How to simplify use of preprocessing wildcard?

[Fedorov113]

Final?
Longest - latest?

[Fedorov113]

Inside Dataset folder we have fs_samples.tsv with columns fs_name final_preproc use_in_anal and arbitrary other columns.
This file is being generated automatically, when we add new samples to the Dataset, and it is being updated when finishing any requested preprocessing Result. final_preproc is set to raw (or other initial preproc name) on import/discovering new sample. By default, preprocessing sets final_preproc to theirs name. Also we need Command set-final-preproc.
Also we can have reserved keywords such as __longest__, __latest__.
fs_samples.tsv is loaded when creating Dataset instance.
We also need sample_containers.tsv, long format table indexed by fs_name + preproc.
If preprocessing Result finished - sample_containers should be updated. Preprocessing Results should leave {sample}.verdict.txt

IMPORTANT: We update fs_samples.tsv and sample_containers.tsv only when Snakemake finishes and we check that everythin is ok in assnake.cli.commands.cmd_run, not on each rule completition.