allow range to satisfy key distribution generally

[gene-bordegaray]

Which issue does this PR close?

• Closes Allow Partitioning::Range to satisfy Distribution::KeyPartitioned generally #23266.
• Part of [EPIC] Implement Range Partitioning #22395.

Rationale for this change

Partitioning::Range now satisfies Distribution::KeyPartitioned privately across all operators that require it: aggregates, windows, TopK, and co-partitioned joins. So now the temporary operator opt-ins can be consolidated.

What changes are included in this PR?

• Allow compatible Partitioning::Range to satisfy Distribution::KeyPartitioned via Partitioning::satisfaction
• Remove the temporary range-satisfaction helpers and operator-specific opt-ins

Are these changes tested?

Yes

Are there any user-facing changes?

No, this should not change any exisitng behavior just consolidation

[codecov-commenter]

Codecov Report

❌ Patch coverage is 98.02632% with 3 lines in your changes missing coverage. Please review.
✅ Project coverage is 80.71%. Comparing base (eef1017) to head (03518df).
⚠️ Report is 13 commits behind head on main.

Files with missing lines	Patch %	Lines
datafusion/physical-expr/src/partitioning.rs	97.84%	0 Missing and 3 partials ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #23680      +/-   ##
==========================================
- Coverage   80.71%   80.71%   -0.01%     
==========================================
  Files        1089     1089              
  Lines      368748   368741       -7     
  Branches   368748   368741       -7     
==========================================
- Hits       297633   297614      -19     
- Misses      53372    53374       +2     
- Partials    17743    17753      +10     

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[gene-bordegaray]

cc: @gabotechs

[gabotechs]

Nice cleanup! 👍

[gabotechs]

I see some tests where deleted here? are they no longer relevant?

[gene-bordegaray]

I just consolidated these into the test_partitioning_key_satisfaction() test. Actually after you pointing this out I think I will rename this test_hash_partitioning_key_satisfaction() and then keep the test_range_partitioning_key_satisfaction()

I got rid of some case that were repearting testing the same code path such as:

"KeyPartitioned([a, b, c]) satisfied by Hash([b, a])"

and 

"KeyPartitioned([a, b, c]) satisfied by Hash([a, b])"

[gene-bordegaray]

were there any that popped out to you as bad to remove?

[gabotechs]

None specifically, just pointing it out so that we don't loose coverage. If we are still covered, then it should be good 👍

[gene-bordegaray]

Sounds good. I will double check when renaming test

[gene-bordegaray]

renamed and went over tests to ensure we didn't lose coverage.

I added back: Hash([unknown]) vs KeyPartitioned([unknown,a]) to check an unknown cant match itself in subset