Replacement of "Header" objects with mth5 and mt_metadata objects

[kkappler]

Jared to give Karl a tutorial on the station level object and some examples.
We also should review how these metadata objects work with MTPy, it is possible that much of the TF manipulation post computation can be done using existing tools. For example any corrections to TF associated with rotations etc.

A good place to try plugging this in maybe
transfer_function_header.TransferFunctionHeader.array_header_to_tf_header(self, array_header, sites):

[kkappler]

this relates to issue #79

[kkappler]

The "station header" and rest of the metadata are being packed into the mt_metadata transfer function container via:

tf_cls.station_metadata.from_dict(run_metadata_dict)
tf_cls.survey_metadata.from_dict(survey_dict)

where we create the metadata dictionaries via:

run_metdata_dict = local_run_obj.station_group.metadata.to_dict()
survey_dict = mth5_obj.survey_group.metadata.to_dict()

There doesn't seem to be a need for the TF calculation to otherwise carry this information during the regression so I suggest that we eliminate the need for a TFHeader class entirely

[kkappler]

Opened Issue #109 for eventual deprecation of this class now that the information is available via mth5 and mt_metadata tf class