feat(python): Add simple version of json_normalize

[ritchie46]

This adds a json_normalize function forked and adapted from pandas. There are some changes, as we will never return an pl.Object type for objects that are not normalized. These will be encoded as json strings, so that we will get a string column and not objects.

Features we have over the pandas counterpart is that we allow the user to set the schema.

Furthermore we don't have all the features, but this should already be useful for the most common cases.

[stinodego]

I would say this isn't technically I/O functionality - it's "convert" functionality since it starts from an in-memory representation. It should be in the convert.py module next to from_arrow, from_numpy, etc.

Also, I'd suggest marking this as unstable for now.

[codecov]

Codecov Report

Attention: Patch coverage is 71.66667% with 17 lines in your changes missing coverage. Please review.

Project coverage is 81.04%. Comparing base (a81f6a5) to head (951b9a4).
Report is 9 commits behind head on main.

Files	Patch %	Lines
py-polars/polars/convert/normalize.py	69.09%	10 Missing and 7 partials ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #17015      +/-   ##
==========================================
+ Coverage   81.03%   81.04%   +0.01%     
==========================================
  Files        1446     1449       +3     
  Lines      190420   190482      +62     
  Branches     2709     2722      +13     
==========================================
+ Hits       154299   154370      +71     
+ Misses      35624    35609      -15     
- Partials      497      503       +6     

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