FE-1222: Classify Petrinaut optimization transport errors in the UI

[kube]

🌟 What is the purpose of this PR?

When a long optimization stream is reset mid-run, the UI shows the raw network error — the caught fetch/reader.read() exception message forwarded verbatim. This replaces it with a classified, actionable error and threads NodeAPI's correlation ids to the UI, so a failed run is traceable to the NodeAPI and optimizer logs and the user gets a clear next step. Already-received trials are preserved and a Retry action is offered.

🔗 Related links

• FE-1222 — this issue
• FE-1217 — parent hardening issue
• SRE-831 — the suspected root cause (proxy/LB streaming timeout); this PR improves the experience regardless of it

🚫 Blocked by

• Stacked on codex/petrinaut-optimization-frontend (FE-1162: Add Petrinaut optimization UI #9051)
• Relies on the X-Optimization-Run-ID response header NodeAPI sets in FE-1221: Add structured correlated logging across the Petrinaut optimization pipeline #9059 for the run-id diagnostic (degrades gracefully to null if absent)

🔍 What does this change?

• messages.ts: optimizationResponseStart carries the x-hash-request-id / X-Optimization-Run-ID response headers; optimizationError gains a category (network / http / protocol / aborted) plus optional correlation ids, httpStatus, and errorName.
• Host (process-editor.tsx): captures the correlation headers, forwards them on response-start, and classifies its own transport failures as network with a safe message (no raw exception text).
• Bridge (create-bridge-petrinaut-optimization.ts): a new PetrinautOptimizationTransportError carries the structured fields; non-ok responses are classified http (with status + correlation ids), protocol violations protocol, and the response-start timeout network; correlation ids captured at response-start backfill a later stream error.
• Provider (provider.tsx): builds a safe, progress-aware message ("Connection to the optimization service was interrupted after N of M trials. Retry the optimization. (diagnostic id: …)"), keeps the received trials, stores errorCategory / errorDiagnostics on the record, and adds retryOptimization to start a fresh run from the failed one's input. The error type crosses the app→lib boundary, so it is duck-typed rather than matched with instanceof.
• View drawer: a Retry action on a failed run.

Pre-Merge Checklist 🚀

🚢 Has this modified a publishable library?

This PR:

• does not modify any publishable blocks or libraries, or modifications do not need publishing

📜 Does this require a change to the docs?

The changes in this PR:

• require changes to docs which are made as part of this PR

🕸️ Does this require a change to the Turbo Graph?

The changes in this PR:

• do not affect the execution graph

🛡 What tests cover this?

• create-bridge-petrinaut-optimization.test.ts: protocol classification (no-terminal, data-after-terminal) and the HTTP path carrying status + correlation ids.
• create-bridge-petrinaut-optimization.browser.test.ts: a mid-stream host error is reclassified and backfilled with the response-start correlation ids.
• provider.test.tsx: a classified transport failure yields the composed message + preserved trials + errorCategory/errorDiagnostics; retryOptimization starts a fresh run from the original input.
• messages.test.ts, create-optimization-drawer.test.tsx: kept green with the extended message and context shapes.

All four suites pass (@hashintel/petrinaut react + UI optimization tests; @apps/hash-frontend bridge + messages tests).

❓ How to test this?

1. Check out the branch / open the deployment.
2. Start a long optimization and interrupt the connection (or let a proxy reset it) so the stream drops mid-run.
3. Confirm the result drawer shows a classified message with the trial progress and a diagnostic id, the already-received trials remain listed, and a Retry button starts a fresh run.

🤖 Generated with Claude Code

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
hash	Ready	Preview, Comment	Jul 19, 2026 4:36pm
petrinaut	Ready	Preview, Comment	Jul 19, 2026 4:36pm

1 Skipped Deployment

Project	Deployment	Actions	Updated (UTC)
hashdotdesign-tokens	Ignored		Jul 19, 2026 4:36pm

[semgrep-code-hashintel]

Semgrep found 1 insufficient-postmessage-origin-validation finding:

• apps/hash-frontend/src/pages/processes/[uuid]/embed.page/create-bridge-petrinaut-optimization.ts
  • L107-213 - Triage

No validation of origin is done by the addEventListener API. It may be possible to exploit this flaw to perform Cross Origin attacks such as Cross-Site Scripting(XSS).