fix(python): Address inconsistency with use of Python types in frame-level cast

[alexander-beedie]

We supported df.cast({"col": str}) but raised an error on df.cast(str); made the behaviour for both styles of call consistent.

Also:
Bonus docstring examples for Schema methods.

Example

import polars as pl

df = pl.DataFrame({
    "colx": [100_000, 200_000, 300_000],
    "coly": [date(1999,12,31), date(2020,1,1), date(2077,7,5)],
})

Before:

df.cast(str)
# TypeError: argument 'dtype': 'type' is not a Polars data type

After:

df.cast(str) 
# shape: (3, 2)
# ┌────────┬────────────┐
# │ colx   ┆ coly       │
# │ ---    ┆ ---        │
# │ str    ┆ str        │
# ╞════════╪════════════╡
# │ 100000 ┆ 1999-12-31 │
# │ 200000 ┆ 2020-01-01 │
# │ 300000 ┆ 2077-07-05 │
# └────────┴────────────┘

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 79.86%. Comparing base (047e578) to head (d88600e).
Report is 21 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #19657      +/-   ##
==========================================
- Coverage   79.92%   79.86%   -0.06%     
==========================================
  Files        1536     1537       +1     
  Lines      211700   211924     +224     
  Branches     2445     2446       +1     
==========================================
+ Hits       169199   169254      +55     
- Misses      41946    42116     +170     
+ Partials      555      554       -1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[stinodego]

Looks good, thanks!