reverted to older code that used an array to store the python unicode values and to support passing a none terminated list or unterminated list by setting an expected length of the input_args before allcocating memory for the c_input_args variable

String comparison must use `==`

See merge request adi/adi_py!4

in correct_zero_length_dims fixed syntax of ADILogger.info and added extra call to ADILogger.info to document dimensions whose lengths are set to 1

Added Krista's fix to correct missing dimensions in dsproc before output datasets are converted to xarray

Added support for setting DQRs

Fixed bug with SkipProcessingInterval exception not logging the message to dsproc.  Fixed bug where we now throw readable exception if trying to convert an adi dataset to xarray and the adi dataset has no time values.

Fixed line endings and fixed a bug where we were calling dsproc.set_sample_timevals with an extra parameter that we didn't need.

Fixed all the get_X_dataset methods to throw an exception if the list of obs is > 1.  The exception tells the user to use the get_X_datasets method instead.
Updated the README file.

Clean up get dataset function signatures

See merge request adi/adi_py!3

Update run_dq_inspector binding

See merge request adi/adi_py!2

Fix `Process.get_datastream_files`

Closes #11

See merge request adi/adi_py!1

Co-authored-by: Eddie Schuman <eddie.schuman@pnnl.gov>

updated fix bug in dsproc.error so thta fi format string is None that the format is set to status as the documentation noted it would.

Added one more bit description to check for old ingest dod templates when picking the default value to use for qc_ variables with empty data.  Updated binding method dsproc.dtype_to_cds_type to also support numpy datetime64 data types (converts to DOUBLE).  Updated README file.

Made some improvements to adi_py methods per feedback from Krista to better support obs-based processes.