chore: Test RcTracker with starting inc_rc elision re-added

[jfecher]

Description

Problem*

Resolves

Summary*

Fork of #6783 to test it on external repos with the change to start reference counts at 2 reverted to start at 1 again.

Additional Context

Documentation*

Check one:

• No documentation needed.
• Documentation included in this PR.
• [For Experimental Features] Documentation to be submitted in a separate PR.

PR Checklist*

• I have tested the changes locally.
• I have formatted the changes with Prettier and/or cargo fmt on default settings.

[github-actions]

Changes to Brillig bytecode sizes

Generated at commit: b1f1d3388052ff162fe979a409dd48a749803002, compared to commit: f597b976a6a324cee9bc327b73d40211be05903e

🧾 Summary (10% most significant diffs)

Program	Brillig opcodes (+/-)	%
simple_shield	-15 ✅	-1.66%
sha256_var_padding_regression	-90 ✅	-1.75%
keccak256	-39 ✅	-1.78%
brillig_cow_assign	-3 ✅	-2.13%
conditional_regression_661	-3 ✅	-2.36%
regression_4202	-3 ✅	-2.40%
fold_call_witness_condition	-3 ✅	-2.61%

---

Full diff report 👇

Program	Brillig opcodes (+/-)	%
u128	2,754 (-3)	-0.11%
slice_dynamic_index	2,504 (-3)	-0.12%
nested_array_dynamic	2,454 (-3)	-0.12%
strings	914 (-3)	-0.33%
array_to_slice	715 (-3)	-0.42%
higher_order_functions	674 (-3)	-0.44%
debug_logs	5,124 (-24)	-0.47%
poseidon_bn254_hash	5,424 (-27)	-0.50%
poseidon_bn254_hash_width_3	5,424 (-27)	-0.50%
regression_6674_3	552 (-3)	-0.54%
poseidonsponge_x5_254	4,284 (-24)	-0.56%
regression	939 (-6)	-0.63%
regression_5252	4,651 (-30)	-0.64%
hashmap	21,662 (-147)	-0.67%
poseidon2	332 (-3)	-0.90%
sha256_brillig_performance_regression	1,647 (-15)	-0.90%
bench_2_to_17	325 (-3)	-0.91%
uhashmap	13,884 (-132)	-0.94%
ecdsa_secp256k1	899 (-9)	-0.99%
array_dynamic_nested_blackbox_input	892 (-9)	-1.00%
wildcard_type	295 (-3)	-1.01%
conditional_1	1,176 (-12)	-1.01%
fold_2_to_17	578 (-6)	-1.03%
regression_struct_array_conditional	570 (-6)	-1.04%
slice_loop	266 (-3)	-1.12%
sha256_regression	6,839 (-78)	-1.13%
array_dynamic_blackbox_input	1,020 (-12)	-1.16%
sha2_byte	2,734 (-33)	-1.19%
regression_6674_2	247 (-3)	-1.20%
regression_6674_1	244 (-3)	-1.21%
brillig_cow_regression	2,140 (-27)	-1.25%
fold_complex_outputs	475 (-6)	-1.25%
reference_counts	467 (-6)	-1.27%
bigint	2,165 (-30)	-1.37%
slices	2,156 (-30)	-1.37%
databus_two_calldata	211 (-3)	-1.40%
to_be_bytes	209 (-3)	-1.42%
conditional_regression_short_circuit	1,246 (-18)	-1.42%
pedersen_commitment	206 (-3)	-1.44%
merkle_insert	793 (-12)	-1.49%
6	1,173 (-18)	-1.51%
ram_blowup_regression	954 (-15)	-1.55%
7_function	570 (-9)	-1.55%
no_predicates_numeric_generic_poseidon	748 (-12)	-1.58%
fold_numeric_generic_poseidon	748 (-12)	-1.58%
pedersen_check	556 (-9)	-1.59%
brillig_pedersen	556 (-9)	-1.59%
regression_4449	739 (-12)	-1.60%
sha256_var_witness_const_regression	1,287 (-21)	-1.61%
pedersen_hash	367 (-6)	-1.61%
sha256	2,369 (-39)	-1.62%
brillig_oracle	364 (-6)	-1.62%
aes128_encrypt	543 (-9)	-1.63%
sha256_var_size_regression	1,787 (-30)	-1.65%
simple_shield	890 (-15)	-1.66%
sha256_var_padding_regression	5,039 (-90)	-1.75%
keccak256	2,158 (-39)	-1.78%
brillig_cow_assign	138 (-3)	-2.13%
conditional_regression_661	124 (-3)	-2.36%
regression_4202	122 (-3)	-2.40%
fold_call_witness_condition	112 (-3)	-2.61%

[github-actions]

Peak Memory Sample

Program	Peak Memory	%
keccak256	79.14M	0%
workspace	121.88M	-1%
regression_4709	295.98M	0%
ram_blowup_regression	2.44G	0%
private-kernel-tail	210.39M	0%
private-kernel-reset	860.67M	-1%
private-kernel-inner	307.77M	0%
parity-root	175.77M	0%

[github-actions]

Compilation Sample

Program	Compilation Time	%
sha256_regression	0m1.506s	4%
regression_4709	0m0.803s	7%
ram_blowup_regression	0m17.480s	3%
private-kernel-tail	0m1.131s	-2%
private-kernel-reset	0m9.351s	2%
private-kernel-inner	0m2.529s	6%
parity-root	0m0.886s	0%
noir-contracts	2m37.104s	-4%

[vezenovm]

Fork of #6783 to test it on external repos with the change to start reference counts at 2 reverted to start at 1 again.

Looks like the first assert in reference_counts needs to be updated

[github-actions]

Changes to number of Brillig opcodes executed

Generated at commit: b1f1d3388052ff162fe979a409dd48a749803002, compared to commit: f597b976a6a324cee9bc327b73d40211be05903e

🧾 Summary (10% most significant diffs)

Program	Brillig opcodes (+/-)	%
brillig_oracle	-318 ✅	-11.56%
pedersen_commitment	-47 ✅	-16.04%
brillig_pedersen	-190 ✅	-19.55%
pedersen_check	-190 ✅	-19.55%
simple_shield	-584 ✅	-19.81%
merkle_insert	-858 ✅	-21.34%
pedersen_hash	-143 ✅	-21.38%

---

Full diff report 👇

Program	Brillig opcodes (+/-)	%
bench_2_to_17	589,838 (-3)	-0.00%
fold_2_to_17	1,093,919 (-6)	-0.00%
slice_dynamic_index	4,664 (-3)	-0.06%
strings	1,767 (-3)	-0.17%
array_to_slice	1,667 (-3)	-0.18%
regression	2,810 (-6)	-0.21%
slice_loop	939 (-3)	-0.32%
regression_6674_2	819 (-3)	-0.36%
regression_6674_1	816 (-3)	-0.37%
poseidon_bn254_hash	165,946 (-689)	-0.41%
poseidon_bn254_hash_width_3	165,946 (-689)	-0.41%
hashmap	55,279 (-237)	-0.43%
poseidon2	695 (-3)	-0.43%
debug_logs	5,136 (-24)	-0.47%
poseidonsponge_x5_254	183,546 (-864)	-0.47%
regression_5252	916,132 (-4,366)	-0.47%
u128	24,831 (-131)	-0.52%
brillig_cow_assign	533 (-3)	-0.56%
uhashmap	146,378 (-930)	-0.63%
fold_complex_outputs	762 (-6)	-0.78%
slices	3,335 (-30)	-0.89%
nested_array_dynamic	3,417 (-47)	-1.36%
regression_4202	211 (-3)	-1.40%
reference_counts	397 (-6)	-1.49%
conditional_regression_661	117 (-3)	-2.50%
sha2_byte	45,276 (-1,435)	-3.07%
higher_order_functions	1,215 (-40)	-3.19%
regression_6674_3	1,561 (-54)	-3.34%
ecdsa_secp256k1	6,540 (-249)	-3.67%
sha256_brillig_performance_regression	22,611 (-904)	-3.84%
fold_numeric_generic_poseidon	4,935 (-205)	-3.99%
no_predicates_numeric_generic_poseidon	4,935 (-205)	-3.99%
sha256_regression	116,354 (-4,918)	-4.06%
sha256_var_size_regression	16,315 (-714)	-4.19%
fold_call_witness_condition	67 (-3)	-4.29%
array_dynamic_nested_blackbox_input	4,301 (-249)	-5.47%
conditional_regression_short_circuit	7,030 (-498)	-6.62%
keccak256	32,474 (-2,303)	-6.62%
6	6,952 (-498)	-6.68%
sha256	13,869 (-1,127)	-7.52%
array_dynamic_blackbox_input	17,524 (-1,432)	-7.55%
regression_4449	207,712 (-17,433)	-7.74%
7_function	2,333 (-204)	-8.04%
aes128_encrypt	4,326 (-393)	-8.33%
conditional_1	5,225 (-492)	-8.61%
sha256_var_witness_const_regression	6,571 (-629)	-8.74%
wildcard_type	471 (-47)	-9.07%
ram_blowup_regression	717,859 (-74,380)	-9.39%
brillig_cow_regression	477,835 (-50,025)	-9.48%
to_be_bytes	2,216 (-236)	-9.62%
regression_struct_array_conditional	1,515 (-164)	-9.77%
sha256_var_padding_regression	204,188 (-22,308)	-9.85%
databus_two_calldata	399 (-47)	-10.54%
brillig_oracle	2,433 (-318)	-11.56%
pedersen_commitment	246 (-47)	-16.04%
brillig_pedersen	782 (-190)	-19.55%
pedersen_check	782 (-190)	-19.55%
simple_shield	2,364 (-584)	-19.81%
merkle_insert	3,162 (-858)	-21.34%
pedersen_hash	526 (-143)	-21.38%

[TomAFrench]

Seems like we're getting correctness issues inside of bigcurve again.

[jfecher]

Those are expected since this should have no changes compared to the breaking version that was removed.
This was mainly to see if we'd run into the failures in aztec packages again with mv/bring-back-rc-tracker and this change