tfs_pandas writer: misleading error-message if two columns have the same name

[JoschD]

The tfs writer crashes with:
AttributeError: dtype is not in the DataFrame or headers.
when two columns have the same name.
If we do not want to allow that in the first place, we should have a check in e.g. _validate()

Example Code:

from utils import tfs_pandas as tfs
df = tfs.TfsDataFrame([1,2,3,4], index=["A"], columns=["B", "B", "C", "D"])
tfs.write_tfs("test.tfs", df)