feat(avm)!: callstack metadata collector (WIP for CI)

[fcarreiro]

TODO/COMMENTS

• Think whether this approach makes sense
• Document/comment caveats and footguns: e.g., you may not find a revertreason if you went beyond limits
• Make tests pass
• Enforce collection limits in C++
• Consider having/plumbing more fine grained errors in C++/gadgets
• generated simulation errors and selectors/function names: can the sequencer get the function name from the DB? or does it depend on artifacts? In any case, looks like in the TS flow we always put a string in "functionName" even if we don't know it. This might be wrong because the enriching fn might assume the name is already there. On C++ we seem to always fill in functionSelector. Is this ok or do we want to DI a contracts db and try to get the name? [this feature is minor right now, except for checking that TS does it right/wrong]

[fcarreiro]

• #18462 : 2 dependent PRs (#18614 , #18696 ) 👈 (View in Graphite)
• merge-train/avm

This stack of pull requests is managed by Graphite. Learn more about stacking.

[AztecBot]

Flakey Tests

🤖 says: This CI run detected 1 tests that failed, but were tolerated due to a .test_patterns.yml entry.

\033FLAKED\033 (\0338;;http://ci.aztec-labs.com/9fc308c9e992282f9fc308c9e992282f8;;\033): yarn-project/end-to-end/scripts/run_test.sh simple src/e2e_p2p/inactivity_slash.test.ts (369s) (code: 1) group:e2e-p2p-epoch-flakes (\033dbanks12\033: feat(avm): callstack metadata collection limits are properly plumbed to cpp simulator)