fix: Fix incorrect output height for SQL SELECT COUNT(*) FROM

[nameexhaustion]

• Fixes Count(*) does not aggregate rows #21084

Note that this fix will also change the behavior of the following query (also adds a test for it):

df = pl.DataFrame(
    {
        "list_long": [[1, 2, 3], [4, 5, 6]],
        "sort_key": [2, 1],
    }
)

df.sql("SELECT UNNEST(list_long) FROM self ORDER BY sort_key")

# Before
polars.exceptions.ShapeError: unable to add a column of length 6 to a DataFrame of height 2
# After
shape: (6, 1)
┌───────────┐
│ list_long │
│ ---       │
│ i64       │
╞═══════════╡
│ 2         │
│ 1         │
│ 3         │
│ 4         │
│ 5         │
│ 6         │
└───────────┘

Drive-by fixes LazyFrame::join() in Rust ignoring the maintain_order option.

[codecov]

Codecov Report

Attention: Patch coverage is 98.93617% with 1 line in your changes missing coverage. Please review.

Project coverage is 79.30%. Comparing base (15fa50a) to head (ae18074).
Report is 2 commits behind head on main.

Files with missing lines	Patch %	Lines
crates/polars-lazy/src/frame/mod.rs	96.77%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #21108      +/-   ##
==========================================
+ Coverage   79.29%   79.30%   +0.01%     
==========================================
  Files        1589     1589              
  Lines      226515   226597      +82     
  Branches     2594     2594              
==========================================
+ Hits       179611   179703      +92     
+ Misses      46308    46298      -10     
  Partials      596      596              

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