docs(python): Add description and example of what happens when ignore_nulls=True on pl.concat_str

[bertiewooster]

Addresses #16814 on https://docs.pola.rs/py-polars/html/reference/expressions/api/polars.concat_str.html#polars.concat_str. (Other than changing the font, which is a larger decision.)

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 80.84%. Comparing base (0b0af39) to head (3865dcb).
Report is 184 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #16889      +/-   ##
==========================================
- Coverage   81.33%   80.84%   -0.49%     
==========================================
  Files        1425     1466      +41     
  Lines      187867   192324    +4457     
  Branches     2702     2745      +43     
==========================================
+ Hits       152797   155481    +2684     
- Misses      34574    36340    +1766     
- Partials      496      503       +7     

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

[bertiewooster]

@DeflateAwning does this address the issue?

[DeflateAwning]

The example looks great! Assuming executing that example works as indicated, I believe your description is actually wrong.

You wrote:

If True, null values will be propagated and will appear to be empty strings.

In reality:

If True, null values are removed from the input list before concatenation.

Issues:

• "Propogated" means "passed up to the caller, thus making the output cell null"
• "Appear to be empty strings" would mean that the delimeter is duplicated.

[bertiewooster]

Good catch; I wasn't thinking about the separator. I corrected the description.
BTW, to pass the doctests, the code example output has to be correct.

[DeflateAwning]

Looks good to me!

[bertiewooster]

Request final review and, if approved, merging on this PR. It was endorsed by the original requestor. Thanks!