feat(prover): improve proof submission delay calculation

[davidtaikocha]

use:

if basefee != 0 {
  target_delay = expected_reward/basefee * target;
  if target_delay < target * 0.25 {
    target_delay = target * 0.25;
  } else if target_delay > target * 4 {
    target_delay = target * 4;
  }
} else {
  target_delay = target * 4;
}

[github-actions]

Overall, this pull request looks good. There are a few minor things I would recommend to improve the code:

• The variable names could be more descriptive and consistent, e.g. targetDelay and stateVar.
• The calculation of the proof submission delay could be more concise.
• Consider using constants for the proofTimeTarget and blockFee values.

[codecov]

Codecov Report

Merging #249 (6133b2c) into main (f64a762) will decrease coverage by 0.03%.
The diff coverage is 20.68%.

@@            Coverage Diff             @@
##             main     #249      +/-   ##
==========================================
- Coverage   53.11%   53.09%   -0.03%     
==========================================
  Files          36       36              
  Lines        3579     3586       +7     
==========================================
+ Hits         1901     1904       +3     
- Misses       1430     1434       +4     
  Partials      248      248              

Impacted Files	Coverage Δ
driver/state/state.go	62.98% <0.00%> (+2.02%)	⬆️
prover/proof_producer/zkevm_cmd_producer.go	23.46% <0.00%> (ø)
prover/proof_producer/zkevm_rpcd_producer.go	27.55% <0.00%> (ø)
prover/prover.go	51.68% <ø> (+0.15%)	⬆️
prover/proof_submitter/util.go	32.96% <25.00%> (-2.10%)	⬇️

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[github-actions]

Overall, this Pull Request looks good. The code changes appear to be well thought out and the commit messages provide enough context for the changes.

Some points to consider:

• The variable startAt is used in multiple places. It would be better to use a more descriptive variable name such as proofSubmissionStartTime.
• The logic for calculating the delay could be improved. It might be worth considering using a different approach such as exponential backoff.
• There are some typos in the commit messages that should be corrected.

[github-actions]

Overall comments:

• This pull request looks to be improving the proof submission delay calculation in the prover.
• The code changes appear to be well written and easy to read.

Specific comments:

• It might be worth considering adding some additional documentation to explain the purpose of this change and how it works.
• In Patch 1/3, Line 97, it looks like there is a typo with the variable needNewProof.
• In Patch 2/3, Line 4, it looks like the improve-prover-wait-time-calculation branch is not included in the list of branches that will trigger this workflow.

[github-actions]

Overall, the code changes in this pull request look good. However, there are a few things I would like to point out:

• The variables startAt, targetDelay, and expectedReward are used multiple times in the code. It would be better to use more descriptive variable names to make the code more readable.
• The comment on line 111 should be updated to include the block ID.
• The log message on line 122 should include the amount of time left to wait for the transaction to submit.

[github-actions]

Overall, this pull request looks good. Here are my comments:

• The variable startAt appears to be used in multiple places. It would be better to use more descriptive names for the variables, such as proofSubmissionStartTime or proofSubmissionStartAt.
• The logs could be improved with more descriptive messages and relevant data such as the blockID.
• The code could be refactored to reduce the duplication of similar code blocks.

[github-actions]

Overall, this pull request looks good. Here are some points I noticed:

• In Patch 1/6, the variable startAt is declared twice and could be combined to one.
• In Patch 2/6, the branch name should be consistent with other branch names.
• In Patch 4/6, there is an unnecessary blockID log in Cancel function.
• In Patch 5/6, the log message in Cancel function can be improved for more clarity.
• In Patch 6/6, the branch name should be consistent with other branch names.

[RogerLamTd]

code matches the intended logic, LGTM