Array api docs

[TomAugspurger]

Closes
#24066

cc @datapythonista @jorisvandenbossche there are a few closely related changes here.

• removed api.scalars. I think it's more useful to present "Here are all pandas extensions", and then sub-divide into array / scalar / dtype. I think arrays are the most prominent of these, so I I've named the file api/arrays.rst
• Moved pd.array to api/arrays.rst
• Moved Categorical from api/series.rst to api/arrays.rst
• Re-written the accessor section and moved / left those in api/series.rst
• Added DatetimeArray (the original point of this PR)

---

I have quite a few warnings to cleanup, but wanted a +1 on the general reorganization before putting more time into it.

[TomAugspurger]

These were all moved down one level. The TOC is now

Accessors
    - dt
    - str
    - cat
    - sparse

Rather than having them all at the top-level of this section.

[TomAugspurger]

The metaprogramming to create all the Int* dtypes is causing issues with sphinx / numpydoc

[codecov]

Codecov Report

Merging #24626 into master will decrease coverage by 49.37%.
The diff coverage is n/a.

@@             Coverage Diff             @@
##           master   #24626       +/-   ##
===========================================
- Coverage   92.38%      43%   -49.38%     
===========================================
  Files         166      166               
  Lines       52395    52395               
===========================================
- Hits        48403    22534    -25869     
- Misses       3992    29861    +25869

Flag	Coverage Δ
#multiple	?
#single	43% <ø> (ø)	⬆️

Impacted Files	Coverage Δ
pandas/io/formats/latex.py	0% <0%> (-100%)	⬇️
pandas/core/categorical.py	0% <0%> (-100%)	⬇️
pandas/io/sas/sas_constants.py	0% <0%> (-100%)	⬇️
pandas/tseries/plotting.py	0% <0%> (-100%)	⬇️
pandas/tseries/converter.py	0% <0%> (-100%)	⬇️
pandas/io/formats/html.py	0% <0%> (-99.37%)	⬇️
pandas/core/groupby/categorical.py	0% <0%> (-95.46%)	⬇️
pandas/io/sas/sas7bdat.py	0% <0%> (-91.17%)	⬇️
pandas/io/sas/sas_xport.py	0% <0%> (-90.15%)	⬇️
pandas/core/tools/numeric.py	10.44% <0%> (-89.56%)	⬇️
... and 124 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 cb31b2b...a5ac93d. Read the comment docs.

[codecov]

Codecov Report

Merging #24626 into master will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master   #24626   +/-   ##
=======================================
  Coverage   92.38%   92.38%           
=======================================
  Files         166      166           
  Lines       52395    52395           
=======================================
  Hits        48403    48403           
  Misses       3992     3992

Flag	Coverage Δ
#multiple	90.8% <ø> (ø)	⬆️
#single	43% <ø> (ø)	⬆️

Impacted Files	Coverage Δ
pandas/core/arrays/array_.py	100% <ø> (ø)	⬆️
pandas/core/generic.py	96.62% <ø> (ø)	⬆️
pandas/core/frame.py	96.92% <ø> (ø)	⬆️
pandas/core/arrays/sparse.py	92.17% <ø> (ø)	⬆️

---

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 cb31b2b...d650fa1. Read the comment docs.

[datapythonista]

Really nice, better organized this way. Didn't review with great detail, but looks good.

I just don't quite understand the hierarchy in arrays.rst. Seems to me that Datetime Data should be under Pandas Arrays and not at the same level. May be I'm just not understanding it well.

Also, I tried to be consistent with the blank lines when I splitted api.rst, and I used:

A title
-------
Not a blank line between the ----- line and this.

Not really important, but in this PR some times blank lines are used, and some times are not.

[TomAugspurger]

You're correct about the heading levels. I'll fix those.

[jreback]

thanks!