REF: Changed ExtensionDtype inheritance

[TomAugspurger]

is_extension_array_dtype(dtype) was incorrect for dtypes that haven't
implemented the new interface yet. This is because they indirectly subclassed
ExtensionDtype.

This PR changes the hierarchy so that PandasExtensionDtype doesn't subclass
ExtensionDtype. As we implement the interface, like Categorical, we'll add
ExtensionDtype as a base class.

Before:

DatetimeTZDtype <- PandasExtensionDtype <- ExtensionDtype (wrong)
CategoricalDtype <- PandasExtensionDtype <- ExtensionDtype (right)

After:

DatetimeTZDtype <- PandasExtensionDtype
                                        \
                                         - _DtypeOpsMixin
                                        /
                   ExtensionDtype ------

CategoricalDtype - PandasExtensionDtype -
                \                        \
                 \                        -_DtypeOpsMixin
                  \                      /
                   ExtensionDtype -------

Once all our extension dtypes have implemented the interface we can go back
to the simple, linear inheritance structure.

[jreback]

can you add a test that asserts is_extension_array (or not) for each of the current pandas extension types (and so should only be True for CategoricalDtype, but false for Interval, Period, Datetimewtz)

[codecov]

Codecov Report

Merging #20363 into master will increase coverage by 0.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master   #20363      +/-   ##
==========================================
+ Coverage   91.78%   91.79%   +0.01%     
==========================================
  Files         152      152              
  Lines       49196    49186      -10     
==========================================
- Hits        45153    45152       -1     
+ Misses       4043     4034       -9

Flag	Coverage Δ
#multiple	90.18% <100%> (+0.01%)	⬆️
#single	41.83% <88.46%> (+0.01%)	⬆️

Impacted Files	Coverage Δ
pandas/core/dtypes/dtypes.py	95.9% <100%> (ø)	⬆️
pandas/core/dtypes/base.py	91.89% <100%> (+0.22%)	⬆️
pandas/core/dtypes/common.py	94.64% <100%> (ø)	⬆️
pandas/core/dtypes/cast.py	87.68% <100%> (ø)	⬆️
pandas/core/internals.py	95.54% <100%> (ø)	⬆️
pandas/core/indexes/datetimes.py	95.64% <0%> (ø)	⬆️
pandas/core/generic.py	95.85% <0%> (ø)	⬆️
pandas/core/series.py	93.84% <0%> (ø)	⬆️
pandas/core/indexes/datetimelike.py	96.72% <0%> (ø)	⬆️
pandas/errors/__init__.py	92.3% <0%> (ø)	⬆️
... and 4 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 87f5654...da2451e. Read the comment docs.

[TomAugspurger]

Added tests for each extension dtype. Will have to move from not to is as we migrate them.