Execute tensor L2 norm with RowFn

[connortsui20]

Summary

• Tracking Issue: Epic: Row-oriented scalar functions #9128
• Depends on: Add tensor row inputs for RowFn #9776

Moves L2Norm to the current RowFn API while preserving its direct arithmetic and null behavior. L2Normalize remains on its handwritten one-pass implementation and is not part of this PR.

Changes

Uses the shared tensor row input for ordinary columns and both constant representations. Tests cover constant encodings, empty and zero-width inputs, nulls, and bitwise agreement with materialized rows.

[codspeed-hq]

Merging this PR will degrade performance by 14.26%

⚠️ Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

⚠️ Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

⚡ 4 improved benchmarks
❌ 33 regressed benchmarks
✅ 2150 untouched benchmarks
⏩ 218 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	compress_alp[f64, (1000, 0.0, 1.0)]	104.8 µs	159 µs	-34.08%
❌	Simulation	compress_alp[f64, (1000, 0.0, 0.25)]	114.3 µs	168.2 µs	-32.05%
❌	Simulation	compress_alp[f64, (1000, 0.0, 0.95)]	114.1 µs	167.2 µs	-31.74%
❌	Simulation	compress_alp[f64, (1000, 0.1, 1.0)]	150.6 µs	211.9 µs	-28.91%
❌	Simulation	compress_alp[f64, (1000, 0.01, 1.0)]	144.5 µs	201.1 µs	-28.14%
❌	Simulation	alp_compress_f64	171.8 µs	237.8 µs	-27.74%
❌	Simulation	compress_alp[f64, (1000, 0.01, 0.25)]	155.2 µs	211.9 µs	-26.73%
❌	Simulation	column_x_extension_constant[32]	170.8 µs	232.6 µs	-26.58%
❌	Simulation	compress_alp[f64, (1000, 0.1, 0.25)]	168.9 µs	228.9 µs	-26.22%
❌	Simulation	column_x_extension_constant[256]	427.2 µs	573.8 µs	-25.55%
❌	Simulation	compress_alp[f64, (1000, 0.1, 0.95)]	177.7 µs	237.3 µs	-25.12%
❌	Simulation	compress_alp[f64, (1000, 0.01, 0.95)]	159.8 µs	213.2 µs	-25.03%
❌	Simulation	column_x_column[256]	88.9 µs	112.1 µs	-20.74%
❌	Simulation	column_x_column[32]	94.6 µs	116.9 µs	-19.1%
❌	Simulation	compress_alp[f32, (1000, 0.0, 1.0)]	51.6 µs	63.4 µs	-18.51%
❌	Simulation	compress_alp[f32, (1000, 0.0, 0.25)]	61 µs	73.3 µs	-16.84%
❌	Simulation	compress_alp[f32, (1000, 0.0, 0.95)]	61 µs	73.1 µs	-16.63%
❌	Simulation	column_x_constant[256]	622.7 µs	742.4 µs	-16.13%
❌	Simulation	compress_alp[f64, (10000, 0.0, 1.0)]	308 µs	361 µs	-14.68%
❌	Simulation	compress_alp[f64, (10000, 0.0, 0.95)]	318.3 µs	370.9 µs	-14.18%
...	...	...	...	...	...

ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing ct/row-fn-tensor-l2-v2 (929ae6d) with ct/row-fn-tensor-rows (5eba5b2)²}

Footnotes

1. 218 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

2. No successful run was found on ct/row-fn-tensor-rows (6ca28c3) during the generation of this report, so 63d5d77 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩