Possibility to subsample input data in pipeline #minor

• new rule create_data_subset added to the pipeline
  • at first the metadata is subsampled by backend/scripts/create_metadata_subset.py
  • afterwards the FASTA file is subsampled by backend/scripts/extract_seqs_from_multifasta.sh accordingly
  • the new rule is executed after the rule filter_and_merge
• added tests for new functionalities
• added subsample parameters in config.yaml and adjusted create_config.py script
  • num_seqs for number restriction
  • last_x_days for time restriction
  • priority to set prio when both conditions can't be met
• added section in README file with parameter explanation