Add support for caching program segment

[JulianGCalderon]

Add support for caching program segment

I benchmarked some transactions and found a performance improvement of 3% when running old transactions (Cairo 0).

• The cache used is really simple (thread local). A proper cache for hint data would have more overhead than what I used.
• In my benchmark I ensured that everything is cached beforehand. This won't happen in an actual execution.

Some limitations:

• This PR is not breaking, but to achieve this I had to expose an unsafe API. See safety notes on the load_cached_program function.

• I had to make the memory public, so that the user is able to cache the program segment. We could add a wrapper type to avoid exposing the memory.

Checklist

• Linked to Github Issue
• Unit tests added
• Integration tests added.
• This change requires new documentation.
  • Documentation has been added/updated.
  • CHANGELOG has been updated.

[github-actions]

Benchmark Results for unmodified programs 🚀

Command	Mean [s]	Min [s]	Max [s]	Relative
base big_factorial	2.134 ± 0.011	2.121	2.160	1.00 ± 0.01
head big_factorial	2.132 ± 0.008	2.123	2.146	1.00

Command	Mean [s]	Min [s]	Max [s]	Relative
base big_fibonacci	2.052 ± 0.010	2.039	2.071	1.00
head big_fibonacci	2.061 ± 0.018	2.044	2.106	1.00 ± 0.01

Command	Mean [s]	Min [s]	Max [s]	Relative
base blake2s_integration_benchmark	7.653 ± 0.022	7.613	7.676	1.00
head blake2s_integration_benchmark	7.653 ± 0.079	7.561	7.764	1.00 ± 0.01

Command	Mean [s]	Min [s]	Max [s]	Relative
base compare_arrays_200000	2.177 ± 0.019	2.154	2.222	1.00
head compare_arrays_200000	2.195 ± 0.006	2.187	2.204	1.01 ± 0.01

Command	Mean [s]	Min [s]	Max [s]	Relative
base dict_integration_benchmark	1.434 ± 0.007	1.421	1.444	1.00 ± 0.01
head dict_integration_benchmark	1.430 ± 0.014	1.415	1.466	1.00

Command	Mean [s]	Min [s]	Max [s]	Relative
base field_arithmetic_get_square_benchmark	1.230 ± 0.005	1.223	1.243	1.01 ± 0.01
head field_arithmetic_get_square_benchmark	1.221 ± 0.007	1.214	1.232	1.00

Command	Mean [s]	Min [s]	Max [s]	Relative
base integration_builtins	7.813 ± 0.161	7.720	8.263	1.00 ± 0.03
head integration_builtins	7.798 ± 0.150	7.666	8.147	1.00

Command	Mean [s]	Min [s]	Max [s]	Relative
base keccak_integration_benchmark	8.015 ± 0.104	7.910	8.239	1.00
head keccak_integration_benchmark	8.060 ± 0.114	7.871	8.247	1.01 ± 0.02

Command	Mean [s]	Min [s]	Max [s]	Relative
base linear_search	2.178 ± 0.032	2.154	2.265	1.00
head linear_search	2.192 ± 0.075	2.153	2.394	1.01 ± 0.04

Command	Mean [s]	Min [s]	Max [s]	Relative
base math_cmp_and_pow_integration_benchmark	1.507 ± 0.008	1.495	1.521	1.00 ± 0.01
head math_cmp_and_pow_integration_benchmark	1.504 ± 0.006	1.498	1.514	1.00

Command	Mean [s]	Min [s]	Max [s]	Relative
base math_integration_benchmark	1.455 ± 0.005	1.446	1.463	1.00 ± 0.01
head math_integration_benchmark	1.450 ± 0.007	1.444	1.467	1.00

Command	Mean [s]	Min [s]	Max [s]	Relative
base memory_integration_benchmark	1.206 ± 0.006	1.195	1.215	1.00
head memory_integration_benchmark	1.212 ± 0.013	1.199	1.241	1.00 ± 0.01

Command	Mean [s]	Min [s]	Max [s]	Relative
base operations_with_data_structures_benchmarks	1.572 ± 0.012	1.560	1.598	1.00 ± 0.01
head operations_with_data_structures_benchmarks	1.564 ± 0.006	1.558	1.578	1.00

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
base pedersen	529.3 ± 1.6	526.6	531.6	1.00
head pedersen	531.0 ± 1.4	529.4	533.1	1.00 ± 0.00

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
base poseidon_integration_benchmark	621.9 ± 2.8	617.2	625.5	1.00
head poseidon_integration_benchmark	625.8 ± 6.4	620.3	642.5	1.01 ± 0.01

Command	Mean [s]	Min [s]	Max [s]	Relative
base secp_integration_benchmark	1.840 ± 0.006	1.836	1.855	1.00 ± 0.01
head secp_integration_benchmark	1.837 ± 0.012	1.824	1.864	1.00

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
base set_integration_benchmark	631.7 ± 2.3	628.6	634.7	1.00 ± 0.00
head set_integration_benchmark	629.2 ± 1.6	626.8	632.2	1.00

Command	Mean [s]	Min [s]	Max [s]	Relative
base uint256_integration_benchmark	4.246 ± 0.023	4.218	4.291	1.00
head uint256_integration_benchmark	4.252 ± 0.031	4.212	4.289	1.00 ± 0.01

[codecov]

Codecov Report

❌ Patch coverage is 41.02564% with 23 lines in your changes missing coverage. Please review.
✅ Project coverage is 96.60%. Comparing base (065c8f4) to head (4a6c52f).

Files with missing lines	Patch %	Lines
vm/src/vm/vm_memory/memory.rs	5.26%	18 Missing ⚠️
vm/src/vm/runners/cairo_runner.rs	75.00%	5 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##            2.x.y    #2221      +/-   ##
==========================================
- Coverage   96.66%   96.60%   -0.06%     
==========================================
  Files         103      103              
  Lines       43646    43681      +35     
==========================================
+ Hits        42191    42199       +8     
- Misses       1455     1482      +27     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.