fix: back off invalid metadata jobs and stop re-mints from wiping token metadata

[rafa-stacks]

A token's job row is re-enqueued as pending on every re-mint (the JSON behind a token URI can be fixed without the contract ever changing). But updateJobStatus cleared retry_after for any non-pending status, so a job that had just been marked invalid was pickable again the instant the next mint landed.

Now a UserError sets retry_after = NOW() + JOB_QUEUE_INVALID_RETRY_AFTER_MS (new env var, default 1h).

[Copilot]

Pull request overview

Adds invalid-job backoff and preserves token metadata during re-mints.

Changes:

• Adds configurable invalid-job retry delay.
• Preserves fetched metadata when re-enqueuing minted tokens.
• Adds queue and SFT re-mint regression tests.

Reviewed changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
src/env.ts	Defines the invalid-job backoff setting.
src/pg/stacks-core-pg-store.ts	Preserves metadata and schedules invalid-job retries.
src/token-processor/queue/job/job.ts	Applies backoff after user errors.
tests/token-queue/job.test.ts	Tests invalid-job backoff.
tests/stacks-core/sft-events.test.ts	Tests metadata preservation after re-minting.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.