Allow running reduce_parent operations on stack allocated parents

[robert3005]

Add indirection logic to support stack allocated array parents in reduce rules.

This lets us avoid heap allocation for commonly used operations that are often
immediately optimised away

[codspeed-hq]

Merging this PR will degrade performance by 21.45%

⚠️ 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.

⚡ 26 improved benchmarks
❌ 11 regressed benchmarks
✅ 1171 untouched benchmarks

⚠️ Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	chunked_bool_canonical_into[(10, 1000)]	793.9 µs	920.3 µs	-13.74%
❌	Simulation	chunked_bool_canonical_into[(100, 100)]	102.4 µs	115.5 µs	-11.32%
❌	Simulation	chunked_opt_bool_canonical_into[(10, 1000)]	911.1 µs	1,038.4 µs	-12.26%
❌	Simulation	chunked_varbinview_canonical_into[(10, 1000)]	1.7 ms	1.9 ms	-10.57%
❌	Simulation	search_index_above_max	268.1 µs	340.3 µs	-21.21%
❌	Simulation	search_index_below_min	264.2 µs	336.4 µs	-21.45%
❌	Simulation	search_index_full_range_random	266.2 µs	338.4 µs	-21.34%
❌	Simulation	search_index_in_range	266.5 µs	338.7 µs	-21.32%
❌	Simulation	search_index_mixed_out_of_range	266.3 µs	338.4 µs	-21.3%
⚡	Simulation	slice_dict_tight_loop[10000]	392.2 µs	328.3 µs	+19.48%
⚡	Simulation	slice_primitive_tight_loop[10000]	195.5 µs	166.2 µs	+17.64%
⚡	Simulation	take_search[(0.005, 0.05)]	168.2 µs	132.1 µs	+27.33%
⚡	Simulation	take_search[(0.005, 0.1)]	320.1 µs	247.6 µs	+29.28%
⚡	Simulation	take_search[(0.005, 0.5)]	1.5 ms	1.2 ms	+30.98%
⚡	Simulation	take_search[(0.005, 1.0)]	3.1 ms	2.3 ms	+31.23%
⚡	Simulation	take_search[(0.01, 0.05)]	179.3 µs	143.2 µs	+25.21%
⚡	Simulation	take_search[(0.01, 0.1)]	341 µs	268.5 µs	+27%
⚡	Simulation	take_search[(0.01, 0.5)]	1.6 ms	1.3 ms	+28.53%
⚡	Simulation	take_search[(0.01, 1.0)]	3.3 ms	2.5 ms	+28.75%
⚡	Simulation	take_search[(0.1, 0.05)]	248.9 µs	212.8 µs	+16.96%
...	...	...	...	...	...

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

---

_{Comparing refactor/parent-ref-stack-allocated (04c7c59) with develop (f3dfa0c)}

[joseph-isaacs]

Some of those benchmarks are doing extra real work

[robert3005]

This is not ready yet, there’s still something wrong with matchers