BUG: Series with datetime-like object ops raises TypeError

[sinhrks]

Code Sample, a copy-pastable example if possible

pd.Series([pd.Timestamp('2015-01-01', tz='US/Eastern'),
           pd.Timestamp('2015-01-01', tz='Asia/Tokyo')]) + pd.Timedelta('1 days')
# TypeError: unsupported operand type(s) for +: 'numpy.ndarray' and 'Timedelta'

pd.Series([pd.Period('2015-01-01', freq='D')]) + pd.Timedelta('1 days')
# TypeError: unsupported operand type(s) for +: 'numpy.ndarray' and 'Timedelta'

Expected Output

ops must be applied to each elem

output of pd.show_versions()

current master.

NOTE: This works:

pd.Series([pd.Timestamp('2015-01-01', tz='US/Eastern'),
           pd.Timestamp('2015-01-01', tz='Asia/Tokyo')]) + pd.tseries.offsets.Day()
#0    2015-01-02 00:00:00-05:00
#1    2015-01-02 00:00:00+09:00
# dtype: object

pd.Series([pd.Period('2015-01-01', freq='D')]) + pd.tseries.offsets.Day()
#0   2015-01-02
# dtype: object

[jreback]

@sinhrks thank you sir!