feat(vc): re-execution before attestation

[Maddiaa0]

Overview

Validators currently perform blind attestations to transaction updates

This pr introduces the notion of a light public processor (really light block builder, we can change the name) which

1. Syncs to the latest
2. Executes the transactions in a proposal on top of the head of the chain
3. Does not run the kernel circuits

For calls which currently run lots of kernels ( nested calls ) this currently yields an extremely large performance increase.

In the current test which has an enqueued + 4 nested calls and multiple state updates block building time goes from 2.2seconds to 400 ms ( granted the archive is not calculated )

Downsides:

• This creates a second code path other than the existing public processor
  • although: In my mind it is alot clearer what is going on in this smaller processor

part of : #8720

Issues created from this work

• Fees accounting - feat(vc): reex: check fees #8962
• Calculating the archive state root - feat(vc): reex: calculate archive #8961

[github-actions]

Changes to public function bytecode sizes

Generated at commit: 92ec4b129173685dfbd379b691ebfa90e6be9b23, compared to commit: e416dd41d688a830012dce87f7fc65dd2ed4ccf8

🧾 Summary (100% most significant diffs)

Program	Bytecode size in bytes (+/-)	%
Spam::public_dispatch	+7,382 ❌	+161.81%

---

Full diff report 👇

Program	Bytecode size in bytes (+/-)	%
Spam::public_dispatch	11,944 (+7,382)	+161.81%

[fcarreiro]

I think this is supposed to be the current header and not the historical one. IIRC public will not have access to historical headers, since that would force nodes to be archivers.

[fcarreiro]

OTOH, I'm pretty sure header is currently unused and could be removed (we only keep it for the potential future HEADERMEMBER). When I added it, it was for interoperability with ACVM: #5592

@dbanks12

[Maddiaa0]

aztec-packages/yarn-project/simulator/src/public/public_processor.ts

Line 57 in 2570b59

const historicalHeader = maybeHistoricalHeader ?? merkleTree.getInitialHeader();

I updated it based on the public processor factory - it instantiates the avm executor with a variable called historical header - should this be updated instead?

[Maddiaa0]

#8990

[Maddiaa0]

superceeded by #9768