ENH: Add fold attribute to DatetimeIndex

[gsheni]

Is your feature request related to a problem?

As a pandas user, I wish I could use pandas to determine which values in a DatetimeIndex are using DST (Daylight Saving Time). This could be useful for a datetime column that recorded values year-round (ex - 1 column with EDT and EST datetimes).

Describe the solution you'd like

• An attribute on DatetimeIndex that is named is_daylight_savings_time or is_dst)

API breaking implications

• None, this is an additive feature

Additional context

• This issue resulted from this discussion:
  • ENH: Allow storing timezone-aware datetimes in a series with a datetime64 dtype #46998 (comment)
• This attribute could return the following (for each datetime value):
  • Value is 0 when DST is not in effect.
  • Value is 1 when DST is in effect.
  • Value is -1 when DST is unknown.

[mroeschke]

I tried implementing this a few years ago, and it was discouraged #22560 (comment) so I am not sure it's likely that this will be implemented

[jreback]

idk i think this could be useful (or we add fold instead maybe is better)

[jbrockmendel]

Following #49684, an easy (but non-performant) way to implement this will be folds = np.array([x.fold for x in self], dtype=bool). For a more performant implementation we'd need to refactor out parts of ints_to_pydatetime.