Do not concatenate empty dataframes

[dachengx]

Otherwise, you will see

/home/xudc/strax/strax/run_selection.py:196: FutureWarning: The behavior of DataFrame concatenation with empty or all-NA entries is deprecated. In a future version, this will no longer exclude empty or all-NA columns when determining the result dtypes. To retain the old behavior, exclude the relevant entries before the concat operation.
  docs = pd.concat([docs, new_docs[~np.in1d(new_docs["name"], docs["name"])]], sort=False)

[coveralls]

coverage: 88.913%. remained the same
when pulling d207e4c on no_merge_empty
into 2ded3e0 on master.