[recursion-v2] adjust base layer transcript observing order

[hero78119]

No description provided.

[Copilot]

Pull request overview

This PR adjusts how the base-layer ZKVM transcript is constructed for per-chip proving/verifying, aligning native prover/verifier behavior with the recursion verifier by moving chip metadata observations into per-chip “fork” transcripts (instead of the main transcript) and making the fork transcript initialization explicit.

Changes:

• Remove (circuit_idx, num_instances) observation from the main transcript and instead bind challenges + fork id + metadata into each per-chip transcript.
• Update chip scheduling so transcript “forking” is a plain clone, with task-specific transcript appends done inside the task closure.
• Update recursion verifier to build per-chip challengers independently (labeled fork) and observe the same challenges/metadata order.

Reviewed changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
ceno_zkvm/src/scheme/verifier.rs	Switch per-chip transcript initialization to BasicTranscript("fork") and move challenge/metadata observations into chip transcripts.
ceno_zkvm/src/scheme/scheduler.rs	Remove implicit task_id appends during transcript cloning; closures now bind task-specific transcript state.
ceno_zkvm/src/scheme/prover.rs	Stop writing chip metadata into the main transcript; bind challenges/metadata in per-task transcripts and use a dedicated fork transcript template.
ceno_recursion/src/zkvm_verifier/verifier.rs	Rework per-chip challenger “forking” to independently initialize and observe alpha/beta + fork id + metadata in verifier order.
ceno_recursion/src/transcript/mod.rs	Remove no-longer-used challenger cloning helpers.