Make conversion from tzone to minutes in hatyan.write_components() flexible

[veenstrajelmer]

The code below raises "AttributeError: 'datetime.timezone' object has no attribute '_minutes'":

import hatyan
import ddlpy
import pytz
import matplotlib.pyplot as plt
plt.close('all')
 
start_dt  = "2022-01-01 00:00:00 +01:00"
end_dt  = "2022-03-31 23:50:00 +01:00"

locations = ddlpy.locations()		#retrieve metainfo about available locations/parameters		

bool_grootheid = locations['Grootheid.Code'].isin(['WATHTE']) # measured waterlevels (not astro)
bool_groepering = locations['Groepering.Code'].isin(['NVT']) # timeseries (not extremes)
bool_hoedanigheid = locations['Hoedanigheid.Code'].isin(['NAP']) # vertical reference, only NAP
locs_ddl = locations.loc[bool_grootheid & bool_groepering & bool_hoedanigheid]

donar_loccode = "VLISSGN"
print("   --> processing station:", donar_loccode)
locs_ddl_one = locs_ddl.loc[donar_loccode]
ddl_df = ddlpy.measurements(locs_ddl_one, start_date=start_dt, end_date=end_dt)
df_meas = hatyan.ddlpy_to_hatyan(ddl_df)
# df_meas = df_meas.tz_convert(pytz.FixedOffset(60))
ts_comp_nfac1_fualltimes0_xfac1_peryear0 = hatyan.analysis(ts=df_meas, const_list='month', nodalfactors=True, fu_alltimes=False, xfac=True, analysis_perperiod=False)
ts_comp_nfac1_fualltimes0_xfac1_peryear0.attrs['grootheid'] = "WATHTE"
ts_comp_nfac1_fualltimes0_xfac1_peryear0.attrs['eenheid'] = "cm"
ts_comp_nfac1_fualltimes0_xfac1_peryear0.attrs['vertref'] = "NAP"
ts_comp_nfac1_fualltimes0_xfac1_peryear0.attrs['station'] = donar_loccode
hatyan.write_components(ts_comp_nfac1_fualltimes0_xfac1_peryear0, filename='components_%.ana'%(donar_loccode))

It can be resolve by converting to a fixed offset timezone, but this is not per se desired. Make this more flexible and add a testcase.