remove `date_as_object` argument from `to_dataframe`

[tswast]

I believe this was added to support #11 and #136, but I'm not convinced that this was the right approach. Instead, what we do for TIMESTAMP columns seems more appropriate -- try to cast to the right data type. If it fails due to out-of-range, then fallback to objects.

[tswast]

Per https://pandas.pydata.org/pandas-docs/stable/user_guide/timeseries.html#generating-ranges-of-timestamps

datetime64[ns] is the expected dtype for dates. This is the dtype that is the result of the pandas.date_range function.

[plamut]

OK, will remove it.

[jimfulton]

@plamut have you started on this yet?

I'm running into this as part of my pandas date type. I'd just as soon take this.

[plamut]

@jimfulton Kind of started, but had to switch to other things. If it's a semi-blocker for you that needs to be resolved now-ish, feel free to take it.