Fix api in erna.hdf_utils

[jebuss]

HI, erna gather_fits is not working with the current version:

erna_gather_fits std_analysis 0.17.2 /gpfs0/fact/processing/erna/datasets/Crab1516_no_data_check_std_analysis_0.17.2.hdf5 --start=2015-8-31 --end=2016-3-19 --source="Crab" --datacheck=/gpfs0/fact/processing/erna/no_datacheck.txt
Found 3761 runs with a total ontime of 260.42 h
Outputfile exists! Overwrite? [y, N]: y
  0%|                                                                      | 0/3761 [00:00<?, ?it/s]
Traceback (most recent call last):
  File "/swdev_nfs/anaconda3/bin/erna_gather_fits", line 11, in <module>
    load_entry_point('erna==0.5.0', 'console_scripts', 'erna_gather_fits')()
  File "/swdev_nfs/anaconda3/lib/python3.6/site-packages/click/core.py", line 722, in __call__
    return self.main(*args, **kwargs)
  File "/swdev_nfs/anaconda3/lib/python3.6/site-packages/click/core.py", line 697, in main
    rv = self.invoke(ctx)
  File "/swdev_nfs/anaconda3/lib/python3.6/site-packages/click/core.py", line 895, in invoke
    return ctx.invoke(self.callback, **ctx.params)
  File "/swdev_nfs/anaconda3/lib/python3.6/site-packages/click/core.py", line 535, in invoke
    return callback(*args, **kwargs)
  File "/swdev_nfs/anaconda3/lib/python3.6/site-packages/erna/scripts/gather_fits.py", line 185, in main
    write_fits_to_hdf5(outputfile, successful_jobs.result_file, mode='a')
  File "/swdev_nfs/anaconda3/lib/python3.6/site-packages/erna/hdf_utils.py", line 88, in write_fits_to_hdf5
    compression=compression,
  File "/swdev_nfs/anaconda3/lib/python3.6/site-packages/fact/io.py", line 318, in initialize_h5py
    dtypes = array.dtype
AttributeError: 'numpy.dtype' object has no attribute 'dtype'Ah

The problem is in erna.hdf_utils line 86

if len(names) > 0:
                    array = recfunctions.append_fields(
                        array,
                        names=names,
                        data=arrays,
                        usemask=False,
                    )

                if not initialized:
                    initialize_h5py(
                        hdf_file,
                        array.dtype,
                        key=key,
                        compression=compression,
                    )

Line 86 should be without .dtype

[maxnoe]

This is fixed, right?