BUG: Month Start Grouper raises NonExistentTimeError in some timezones

[mick88]

Pandas version checks

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of pandas.

• I have confirmed this bug exists on the main branch of pandas.

Reproducible Example

import pandas as pd
idx = pd.date_range(start='2017-09-10', end='2017-11-01', freq='1D')
idx = idx.tz_localize('UTC').tz_convert('America/Asuncion')
df = pd.DataFrame(data=list(range(len(idx))), index=idx)
grouper = pd.Grouper(freq='MS')
groups = df.groupby(grouper)

Issue Description

Month Start Grouper raises NonExistentTimeError in some timezones where DST starts on first dat of the month at midnight.

For example in America/Asuncion DST starts on 1st October at midnight.
Running the above example raises the following error:

pytz.exceptions.NonExistentTimeError: 2017-10-01 00:00:00

Available workarounds:
Convert dates to UTC and operate on UTC timezone - this will prevent the error, but will cause dates to be bucketed in the wrong month if datetimes are close to the month start/end. The further the tz is from UTC, the more noticeable the problem.

Expected Behavior

I would expect Month Start to use a valid date, like 2017-10-01 01:00:00 as the start of the month.
nonexistent='shift_forward' argument to Grouper would be appreciated to control this behaviour.

Installed Versions

INSTALLED VERSIONS

commit : e86ed37
python : 3.10.4.final.0
python-bits : 64
OS : Windows
OS-release : 10
Version : 10.0.22621
machine : AMD64
processor : Intel64 Family 6 Model 186 Stepping 2, GenuineIntel
byteorder : little
LC_ALL : None
LANG : None
LOCALE : English_Ireland.1252

pandas : 2.1.1
numpy : 1.22.4
pytz : 2023.3.post1
dateutil : 2.8.2
setuptools : 68.2.2
pip : 22.3.1
Cython : None
pytest : None
hypothesis : None
sphinx : 6.2.1
blosc : None
feather : None
xlsxwriter : 3.1.2
lxml.etree : 4.9.1
html5lib : 1.1
pymysql : None
psycopg2 : 2.9.3
jinja2 : 3.1.2
IPython : 8.16.1
pandas_datareader : None
bs4 : 4.11.1
bottleneck : None
dataframe-api-compat: None
fastparquet : None
fsspec : None
gcsfs : None
matplotlib : None
numba : None
numexpr : None
odfpy : None
openpyxl : 3.0.10
pandas_gbq : None
pyarrow : None
pyreadstat : None
pyxlsb : None
s3fs : None
scipy : None
sqlalchemy : None
tables : None
tabulate : 0.9.0
xarray : None
xlrd : 2.0.1
zstandard : None
tzdata : 2023.3
qtpy : None
pyqt5 : None

[gupta-paras]

take

[mroeschke]

Thanks for the report. I think this is the same core issue as #51211 so closing in favor of that one #51211