Consider catching empty measurement dataframes

[veenstrajelmer]

When calling kw.calc_havengetallen() with an empty dataframe like so:

import pandas as pd
import kenmerkendewaarden as kw
time_index = pd.DatetimeIndex([], dtype='datetime64[ns, Etc/GMT-1]', name='time', freq=None)
df_ext = pd.DataFrame({"HWLWcode":[]},index=time_index)
df_ext.attrs["station"] = "dummy"
kw.calc_havengetallen(df_ext=df_ext)

We get a confusing error from raise_extremes_with_aggers(): "ValueError: df_ext should only contain extremes (HWLWcode 1/2), but it also contains aggers (HWLWcode 3/4/5). You can convert with hatyan.calc_HWLW12345to12()". When fixing this, we get "ValueError: cannot convert float NaN to integer" from kw.havengetallen.calc_HWLW_moonculm_combi(). Consider catching empty dataframes (both wl and ext) troughout kenmerkendewaarden code.

[veenstrajelmer]

Prevented this issue in raise_extremes_with_aggers() by improving it, but since empty dataframes raise different errors throughout the code, one should just not work with them. Does not make sense to add checks to each function.