Cannot read previously reported tuple when using file backend

[donaldcampbelljr]

If you report a tuple to a file backend and then try to re-use that same results file (to report more results), pipestat will fail (well, the yaml constructor.py library throws an error) during the loading of the result files:

yaml.constructor.ConstructorError: could not determine a constructor for the tag 'tag:yaml.org,2002:python/tuple'
  in "<unicode string>", line 8, column 17:
          bed_type: !!python/tuple

How my data appears (it is human readable, I would think it should be machine readable):

default_pipeline_name:
  project: {}
  sample:
    95900d67ed6411a322af35098e445eb0:
      meta:
        pipestat_modified_time: '2025-01-31 12:52:29'
        pipestat_created_time: '2025-01-31 12:52:29'
      bed_type: !!python/tuple
      - bed6+4
      - narrowpeak
    936c62bee58d048eff754c7c189e3d8f:
      meta:
        pipestat_modified_time: '2025-01-31 12:52:29'
        pipestat_created_time: '2025-01-31 12:52:29'
      bed_type: !!python/tuple
      - bed6+4
      - narrowpeak

Obviously, a workaround would be to simply take the values from the tuple and place them in some other form,e.g. two separate strings or a list.

However, I wanted this reporting to be as simple and convenient as possible.