feat(ENG-5578): add integration tests with recorded market replay

[tvinagre]

Summary

• Add record-market CLI tool that captures Tycho protocol stream updates and generates golden quote outputs by replaying through the full Fynd pipeline
• Add integration test suite (fynd-core --test integration) that replays recorded fixtures and verifies solution availability, quality baselines, derived data, and timing
• Worker pools and algorithm config loaded from worker_pools.toml to match production behavior
• Golden file stores exact derived data metrics (spot prices, pool depths, token prices counts) for deterministic replay assertions
• Shared golden types (GoldenFile, TestScenario, etc.) in fynd_core::recording::golden
• parse_chain moved to fynd_core::types::constants for reuse across crates
• Gas price captured from RPC as raw wei string (no lossy float conversion)
• VM-backed protocol states (UniswapV4) filtered during recording serialization
• Git LFS configured for compressed fixture files

Test plan

• cargo +nightly fmt -- --check passes
• cargo +nightly clippy --locked --all --all-features --all-targets -- -D warnings passes
• cargo nextest run --workspace --locked --all-targets --all-features --bin fynd — 416 passed, 0 failed
• All 8 integration tests pass (41/44 scenarios successful, 3 BNB pairs expected NoRouteFound)
• CI runs on GitHub Actions

Generated with Claude Code

[github-actions]

Breaking API Changes (Intentional)

Breaking API changes detected and declared in the PR title.
Ensure the minor version is bumped before merging (breaking changes on 0.x.x bump the minor).

semver-checks output

Checking fynd-core v0.78.1 -> v0.78.1 (no change; assume minor)
     Checked [   0.965s] 196 checks: 196 pass, 49 skip
     Summary no semver update required
    Finished [   1.267s] fynd-core
    Checking fynd-rpc v0.78.1 -> v0.78.1 (no change; assume minor)
     Checked [   0.111s] 196 checks: 195 pass, 1 fail, 0 warn, 49 skip

--- failure function_missing: pub fn removed or renamed ---

Description:
A publicly-visible function cannot be imported by its prior path. A `pub use` may have been removed, or the function itself may have been renamed or removed entirely.
        ref: https://doc.rust-lang.org/cargo/reference/semver.html#item-remove
       impl: https://github.com/obi1kenobi/cargo-semver-checks/tree/v0.46.0/src/lints/function_missing.ron

Failed in:
  function fynd_rpc::builder::parse_chain, previously in file fynd-rpc/src/builder.rs:390

     Summary semver requires new major version: 1 major and 0 minor checks failed
    Finished [   0.175s] fynd-rpc
    Checking fynd-rpc-types v0.78.1 -> v0.78.1 (no change; assume minor)
     Checked [   0.190s] 196 checks: 196 pass, 49 skip
     Summary no semver update required
    Finished [   0.288s] fynd-rpc-types
    Checking fynd-client v0.78.1 -> v0.78.1 (no change; assume minor)
     Checked [   0.238s] 196 checks: 196 pass, 49 skip
     Summary no semver update required
    Finished [   0.375s] fynd-client
    Checking erc20-overrides v0.78.1 -> v0.78.1 (no change; assume minor)
     Checked [   0.084s] 196 checks: 196 pass, 49 skip
     Summary no semver update required
    Finished [   0.125s] erc20-overrides

[kayibal]

Partial review for now. Mostly concerned about the newly exposed api surface from this PR. I think a lot of this API would be better suited to be placed within tycho-simulation if exposed publically.

[tamaralipows]

Thanks - some questions mostly for my understanding

[dianacarvalho1]

Thank you @tvinagre ! I gave a quick review - there are already a lot of good suggestions (that I agree 👍🏼)!

[louise-poole]

Only saw 2 smallish things - otherwise looks good to me!

[louise-poole]

There are a lot of groups multiple tests that call this and with a potential 120s wait each... we risk making our tests excessively slow. Could we maybe reuse the same test harness in all tests/one per test group? Initialise it only once?

[tvinagre]

Good instinct, but the 120s here is a poll deadline, not a wait — wait_until_ready returns as soon as derived data is ready, which is ~15-20s in practice (you can see it in the per-test durations). Reuse isn't really available to us either: nextest runs each test in its own process, so a shared/static harness can never be hit twice, and the Solver's background tasks bind to each test's tokio runtime so sharing would be unsound even under plain cargo test. The one thing that would cut builds is collapsing the tests into fewer #[tokio::test] fns, which we'd rather not do (loses per-test granularity).

The saving grace is that the builds run in parallel: on the last CI run the whole Integration Tests job took 128s vs 216s for Compile & Test, so the suite isn't on the critical path. Happy to revisit (e.g. one harness per logical group via a shared binary) if it grows enough to become the bottleneck.

[tamaralipows]

Perfect - thank you! Nothing else to say here other than Louise's comments