FC sample_rate_decimation_level gives inconsistent values

[kkappler]

Here is a screenshot of showing decimation_level metadata and channel summary dataframe having two different values for
sample_rate_decimation_level.

in mt_metadata/transfer_functions/processing/fourier_coefficients/standards/fc_channel.json this field is described as
"Sample rate of the decimation level."

Should probably

• Update the metadata description to say:
  "Sample rate of the time series that was Fourier transformed to generate the FC decimation level."
  after agreement that this is what is intended.
• Review the methods that generate this value and repair. It looks like the bogus value it is coming from max frequency of the FC array *2, but this is not general beacause in general we have no guarantee that frequencies up to and including the Nyquist will be stored in the FC container. The relevant py files are in mth5/groups/
  • fourier_coefficients.py
  • fc_dataset.py
    Probably best if the value in the dataframe were taken from the metadata (and never computed). Hopefully this make a singe place where the info is sourced.
• Add a test that at least confirms that dec_level.metadata.sample_rate_decimation_level and the fc_level channel_summary df column agree.