Use im instead of conj in compute_fri_input.

[dancarmoz]

Note that conj is not currently used anywhere else.

---

Note

Medium Risk
Touches core verifier arithmetic used to construct FRI inputs; small but correctness-critical changes could invalidate proofs or break interoperability if the algebra doesn’t exactly match the prover.

Overview
Updates the STARK verifier’s compute_fri_input/OodsPointAuxiliary batching math to use the twisted imaginary part im(z) (instead of z - conj(z) / conj(z)), simplifying accumulators and changing the a/b/c/d/e coefficient definitions used in the rational function.

Adds a new circuit op im (with tests) and rewires oods.rs to depend on it, and updates privacy verifier constant hashes to match the new circuit/verifier constants.

^{Written by Cursor Bugbot for commit 9fe2f46. This will update automatically on new commits. Configure here.}

[reviewable-StarkWare]

This change is 

[leo-starkware]

@leo-starkware made 1 comment.
Reviewable status: 0 of 4 files reviewed, 1 unresolved discussion (waiting on dancarmoz and ilyalesokhin-starkware).

---

crates/stark_verifier/src/oods.rs line 260 at r2 (raw file):

}

/// A struct for handling auxiliary values related OODS data corresponding to a point `(px, py)`.

related to

Code quote:

related OODS

[dancarmoz]

@dancarmoz made 1 comment and resolved 1 discussion.
Reviewable status: 0 of 4 files reviewed, all discussions resolved (waiting on ilyalesokhin-starkware).

---

crates/stark_verifier/src/oods.rs line 260 at r2 (raw file):

Previously, leo-starkware wrote…

related to

Done

[leo-starkware]

@leo-starkware reviewed 4 files and all commit messages, and made 1 comment.
Reviewable status: complete! all files reviewed, all discussions resolved (waiting on ilyalesokhin-starkware).