feat: add sync raw send transaction support

[0xl3on]

Motivation

Closes #429

Solution

• Added --send-raw-tx-sync flag to spam and campaign commands
• When enabled, txs are sent via a raw reqwest POST instead of alloy's provider, so we can capture
  end_timestamp_ms right when the response arrives
• Batching (--rpc-batch-size) is disabled when sync is on, we need individual HTTP responses per tx to get per-tx
  timestamps
• For sync txs, process_block_receipts uses the recorded timestamp; normal txs fall back to block header timestamp
  (existing behavior, unchanged)
• About the known limitations: maybe we can start drafting issues about improving the CSV reports timestamp precision

PR Checklist

• Added Tests
• Added Documentation
• Ran cargo +nightly clippy --workspace --lib --examples --tests --benches --all-features --locked --fix
• Ran cargo fmt --all
• Note breaking changes in PR description, if applicable
• update changelogs
  • Update CHANGELOG.md in each affected crate
  • add a high-level description in the root changelog

[0xl3on]

Hey @zeroXbrock ! Closing one more stale issue. Lmk your thoughts!

[zeroXbrock]

Thanks @0xl3on! I'm OOO for the next couple weeks so I'll ask someone from my team if they can review. But if not I'll review it when I'm back!

[zeroXbrock]

nice, works as expected!
Just one thing I'd like to change though: the call to eth_sendRawTransactionSync doesn't show up in reports (neither on the eth_sendRawTransaction Latency graph or the latency quantile table). Naturally, the eth_sendRawTransaction chart renders empty because we didn't call that method, but it would be nice to add a chart for eth_sendRawTransactionSync that takes the place of the eth_sendRawTransaction Latency chart when applicable. Charts with no data should simply not render in the report, and the latency chart should include every method called during the run.

[zeroXbrock]

if these vars need to be read outside of the crate, we should probably just remove these getter functions and make the params public