Enable TopK per Partition (`datafusion.optimizer.enable_window_topn`) by default

[alamb]

Is your feature request related to a problem or challenge?

While writing the DataFusion 55.0.0 blog post

• Blog post for the DataFusion 55.0.0 release #24216
• Add DataFusion 55.0.0 release blog post datafusion-site#203

I wrote up the per-partition TopK window optimization as one of the headline performance improvements of the release — and only then discovered that datafusion.optimizer.enable_window_topn defaults to false, so nobody actually gets it unless they opt in.

We probably shouldn't make a big deal about a feature that is off by default, and I could not find any issue tracking turning it on. So I am filing this issue so we don't forget and we can track what is required to do so in one place rather than scattered across PR review threads.

The story we want to be able to tell

This is the text I had drafted for the 55.0.0 blog post before pulling it, included so it's clear what we're aiming for and so whoever finishes this work can reuse it:

Per-Partition TopK for Window Functions

A common analytics pattern selects the top N rows per group using a window
function:

SELECT * FROM (
    SELECT ROW_NUMBER() OVER (PARTITION BY category ORDER BY revenue DESC) AS rn, *
    FROM sales
) WHERE rn <= 5;

DataFusion previously sorted the entire input to evaluate the window
function, even though only a handful of rows per partition survive the filter.
DataFusion 55 recognizes this pattern and uses a new PartitionedTopKExec
operator that keeps only the top N rows per partition, dramatically reducing
sorting and memory for high-cardinality inputs. The optimization applies to
ROW_NUMBER and RANK, resolving a feature request first filed in 2023
(#6899).
Thanks to @SubhamSinghal for implementing this feature, with reviews from
@2010YOUY01 and @kosiew. Related PRs: #21479, #22885, #23096

Why it is currently off

datafusion/common/src/config.rs:

datafusion/datafusion/common/src/config.rs

Lines 1574 to 1581 in fdf7935

	/// When set to true, the optimizer will replace
	/// Filter(rn<=K) → Window(ROW_NUMBER) → Sort patterns with a
	/// PartitionedTopKExec that maintains per-partition heaps, avoiding
	/// a full sort of the input.
	/// When the window partition key has low cardinality, enabling this optimization
	/// can improve performance. However, for high cardinality keys, it may
	/// cause regressions in both memory usage and runtime.
	pub enable_window_topn: bool, default = false

The default was set to false in #21479 because of severe high-cardinality regressions against the plain-sort baseline:

If it has regressions as large as 0.03x it should [be] off by default (and
we should look if we can automatically enable it via a heuristic / stats based
on partition cardinality / rows)
— @Dandandan, #21479 (comment)

#23096 (in 55.0.0) closed most of that gap by sharing the encoder and memory
reservation across partitions:

Partitions	main	with #23096	vs sort baseline
100	110 ms	105 ms	~1.0x
1,000	117 ms	110 ms	~1.0x
10,000	640 ms	137 ms	1.7x faster than sort (was a regression)
100,000	4,327 ms	320 ms	320 ms vs 238 ms — still slower

but explicitly did not flip the default:

enable_window_topn default stays false per the #21479 discussion — 100K+
remains slower than sort on average, so this PR doesn't motivate flipping the
default. It's the prerequisite for further optimizations that would attack the
residual 100K+ cliff.
— #23096

Describe the solution you'd like

Set enable_window_topn to true by default, once we are confident it does not
regress. Known work items:

• Close the high-cardinality cliff. -- @Dandandan's original suggestion in Perf: Window topn optimisation #21479.
• Fix RANK memory blowup — fix: reduce RANK window top-K memory from O(input) to O(K + ties) per partition #24591
• Fix wide-payload OOM — [EPIC] Memory-efficient dedup on wide payload: fix ROW_NUMBER()=1 and DISTINCT ON patterns #23600
• Expose metrics — PartitionedTopKExec does not expose any metrics (metrics() returns None) #24470 / fix: expose PartitionedTopKExec metrics #24495
• Broaden pattern coverage — Support FilterExec with embedded projections in WindowTopN optimization #21596 / feat(physical-optimizer): support FilterExec with embedded projection in WindowTopN #23599
• Benchmark coverage in CI: WindowTopN dense_rank benchmark #24050 added a dense_rank benchmark; we should make sure the sweep covers high partition-cardinality as well
• Flip the default, run the full benchmark suite, and update docs/source/user-guide/configs.md.

Describe alternatives you've considered

Leave it opt-in indefinitely. 👎 -- an optimization nobody turns on is close to an optimization that doesn't exist

Additional context

• enable_window_topn panics on WHERE rn < 1 (PartitionedTopK requires k > 0) #24404 (panic on WHERE rn < 1) was fixed in fix: skip WindowTopN rewrite when fetch is 0 to avoid PartitionedTopK panic #24405, but that landed after the branch-55 cut
• Optimize "per partition" top-k : ROW_NUMBER < 5 / TopK #6899 (original 2023 request),
• Perf: Window topn optimisation #21479, perf: Extend WindowTopN to support RANK #22885, perf: share encoder/reservation across PartitionedTopKExec partition … #23096, Allow Range partitioned inputs to PartitionedTopK #23355, refactor: moving WindowTopN before EnsureRequirements #24191, [EPIC] Memory-efficient dedup on wide payload: fix ROW_NUMBER()=1 and DISTINCT ON patterns #23600, PartitionedTopKExec does not expose any metrics (metrics() returns None) #24470, fix: reduce RANK window top-K memory from O(input) to O(K + ties) per partition #24591.