merge_coco doesn't work with default setting drop_duplicates=True

[AndreaPi]

Describe the bug
merge_coco doesn't work, because the drop_duplicates code is buggy. Calling it with an exploded list of datasets results in the following error message:

train = merge_coco(*train_folds)
[..]/lib/python3.10/site-packages/cocohelper/dataframe.py:37: UserWarning: COCODataFrame created by a COCODataFrame without using a copy constructor.
  warnings.warn("COCODataFrame created by a COCODataFrame without using a copy constructor.")
[..]/lib/python3.10/site-packages/cocohelper/dataframe.py:37: UserWarning: COCODataFrame created by a COCODataFrame without using a copy constructor.
  warnings.warn("COCODataFrame created by a COCODataFrame without using a copy constructor.")
Traceback (most recent call last):
  File "[..]/pipeline/6a_split_dataset_with_kfold_CV.py", line 29, in <module>
    train = merge_coco(*train_folds)
  File "[..]/lib/python3.10/site-packages/cocohelper/merge.py", line 38, in merge_coco
    return merged.drop_duplicate_cats().drop_duplicate_imgs().drop_duplicate_anns().drop_duplicate_licenses()
  File "[..]/lib/python3.10/site-packages/cocohelper/helper.py", line 743, in drop_duplicate_licenses
    lic_df, id_mapping = drop_duplicate_rows(self.licenses)
  File "[..]/lib/python3.10/site-packages/cocohelper/utils/dataframe.py", line 74, in drop_duplicate_rows
    raise ValueError("There are no columns that can be used to check for duplicates.")
ValueError: There are no columns that can be used to check for duplicates.

To Reproduce
Steps to reproduce the behavior:

1. Create a list of COCOHelper objects, for example by splitting a COCOHelper ch object using KFoldSplitter:

splitter = KFoldSplitter(n_fold=5)
splits = splitter.apply(ch)

2. Try to merge the objects back using merge_coco:

train = merge_coco(*splits)

3. get the error above

Expected behavior
merge_coco should merge the COCOHelper objects without erroring out.

Screenshots
N/A

Desktop (please complete the following information):

  System Version: macOS 13.4 (22F66)
  Kernel Version: Darwin 22.5.0

Additional context
N/A