feat(python): Additional support for loading numpy.float16 values (as Float32)

[alexander-beedie]

Adds an entry to the numpy kind/size conversion mapping covering np.float16.
(We don't have a native Float16, so it loads as Float32).

Example

import numpy as np
import polars as pl

df = pl.DataFrame({
  "n": [np.float16(123)],
})

Before:

# ValueError:
#   cannot parse numpy data type dtype('float16') into Polars data type

After:

# shape: (1, 1)
# ┌───────┐
# │ n     │
# │ ---   │
# │ f32   │
# ╞═══════╡
# │ 123.0 │
# └───────┘

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 79.74%. Comparing base (725c960) to head (cac44fc).
Report is 6 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #20769      +/-   ##
==========================================
- Coverage   79.74%   79.74%   -0.01%     
==========================================
  Files        1561     1561              
  Lines      222012   222012              
  Branches     2530     2530              
==========================================
- Hits       177049   177045       -4     
- Misses      44381    44386       +5     
+ Partials      582      581       -1     

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