[Python] support dateutil timezones

[asfimport]

Here are two main reasons:

1. As of Python 3.6, the tzinfo documentation  recommends dateutil.tz rather than pytz as an IANA time zone provider.

2. Pandas supports dateutil timezones.

   When having a pandas DataFrame that uses a dateutil timezone,  you get an error. 

   Below is a code sample:

   import dateutil
   tz = dateutil.tz.gettz('Asia/Shanghai')
   df = pd.DataFrame({'a': list(range(1, 4)), 'b': pd.date_range('20130101', periods=3, tz=tz)})
   df.to_feather('df.feather')

   Errors:

   ArrowInvalid: ('Object returned by tzinfo.utcoffset(None) is not an instance of datetime.timedelta', "Conversion failed for column b with type datetime64[ns, tzfile('/usr/share/zoneinfo/Asia/Shanghai')]")

    

Reporter: Eric Du

Related issues:

• [Python] support zoneinfo / dateutil timezones (is duplicated by)

_{Note: This issue was originally created as ARROW-10936. Please see the migration documentation for further details.}

[asfimport]

Joris Van den Bossche / @jorisvandenbossche:
[~ericdu] Thanks for the report. This is indeed something we should add.

We already have ARROW-5248 about this as well, so closing this issue as a duplicate