VarBinViewBuilder: support decoding Dict and Sparse arrays directly into the builder

[robert3005]

We can avoid copies if we support scatter and gather operations on builder which
is essentially what dict and sparse array append_to_builder should be

---

_{Stack created with GitHub Stacks CLI • Give Feedback 💬}

[codspeed-hq]

Merging this PR will regress 3 benchmarks

⚠️ 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

⚡ 21 improved benchmarks
❌ 3 regressed benchmarks
✅ 1910 untouched benchmarks
⏩ 51 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	fsst_decompress_string	7.2 ms	8.3 ms	-12.88%
❌	Simulation	decompress_fsst[(10000, 4, 4)]	793.9 µs	899.8 µs	-11.76%
❌	Simulation	decompress_fsst[(10000, 4, 8)]	815.4 µs	920.9 µs	-11.46%
⚡	Simulation	decompress[u64, (1000, 16)]	134.7 µs	73.9 µs	+82.1%
⚡	Simulation	chunked_varbin_to_varbinview_builder[(10, 100)]	516.5 µs	331.5 µs	+55.8%
⚡	Simulation	chunked_varbin_to_varbinview_builder[(500, 2)]	167.6 µs	116.8 µs	+43.53%
⚡	Simulation	chunked_varbin_into_canonical[(10, 100)]	696.3 µs	489.5 µs	+42.25%
⚡	Simulation	chunked_varbin_opt_to_varbinview_builder[(10, 100)]	790.9 µs	568.7 µs	+39.07%
⚡	Simulation	chunked_varbin_into_canonical[(500, 2)]	178.4 µs	128.6 µs	+38.71%
⚡	Simulation	compact_sliced[(16384, 10)]	2 ms	1.4 ms	+37.15%
⚡	Simulation	chunked_varbin_opt_to_varbinview_builder[(500, 2)]	166.2 µs	121.4 µs	+36.83%
⚡	Simulation	compact_sliced[(4096, 10)]	517.9 µs	385.7 µs	+34.29%
⚡	Simulation	chunked_canonicalize_into[(1000, 50, 8, 4)]	15.4 ms	12.8 ms	+20.33%
⚡	Simulation	chunked_canonicalize_into[(1000, 50, 8, 16)]	16.5 ms	13.9 ms	+18.51%
⚡	Simulation	chunked_into_canonical[(1000, 50, 8, 4)]	17.7 ms	15.1 ms	+16.86%
⚡	Simulation	chunked_into_canonical[(1000, 50, 8, 16)]	18.8 ms	16.2 ms	+16.18%
⚡	Simulation	chunked_into_canonical[(1000, 50, 8, 64)]	20.4 ms	17.7 ms	+15.15%
⚡	Simulation	chunked_canonicalize_into[(1000, 50, 8, 64)]	17.8 ms	15.5 ms	+14.73%
⚡	Simulation	take[small_m/shuffled/primitive/nonnull/chunks=16384/indices=16]	1.3 ms	1.1 ms	+13%
⚡	Simulation	take[small_m/shuffled/primitive/nonnull/chunks=16384/indices=256]	3.7 ms	3.4 ms	+11.54%
...	...	...	...	...	...

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 rk/varbinview-gather-scatter (f442d48) with develop (ad5de22)²}

Footnotes

1. 51 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 develop (39fde8c) during the generation of this report, so ad5de22 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩