Expose independently comparable typed result recovery evidence

[flyingrobots]

Outcome

Expose schema-neutral result evidence from original execution, fresh-host recovery, and recovered WAL outcome so an external application can independently compare the exact compiler-authored typed result across recovery.

Acceptance criteria

• The generic runner reports the original, fresh-host recovered, and WAL-recovered application-result evidence as typed schema-neutral fields.
• Each report carries the projection identity, output type, canonical bytes, and result identity already admitted by Echo.
• A passing witness requires all three reports to be byte-identical.
• Negative runner coverage proves substituted recovery or WAL result evidence cannot produce a passing witness.
• No Hello Echo vocabulary, native callback, or application-specific decoder enters Echo production code.

Downstream

Blocks flyingrobots/hello-echo#18.

[coderabbitai]

🔗 Related PRs

#609 - feat(warp-core): project WAL recovery evidence [merged]
#683 - Execute admitted data-only operations in Echo [merged]
#695 - Run compiler-produced Edict operations through durable Ticks [merged]
#700 - Expose duplicate state evidence from the generic runner [merged]
#701 - Evaluate package-declared typed application results [open]

---

📝 Issue Planner

_{Check the box below or use the @coderabbitai plan command to generate an implementation plan and prompts that you can use with your favorite coding assistant.}

• Create Plan

---

🧪 Issue enrichment is currently in open beta.

You can configure auto-planning by selecting labels in the issue_enrichment configuration.

To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

issue_enrichment:
  auto_enrich:
    enabled: false

💬 Have feedback or questions? Drop into our discord!