fix: staging email not delivered

[amandazhuyilan]

Description

Fix: Scheduler crashes after sending emails (staging emails not delivered)

Problem

The email scheduler was silently dying after attempting to send any email. This caused all subsequent emails to never be processed.

Changes

• scheduled_tasks/scheduler.py— Added a dedicated "email", MemoryJobStore alongside the existing SQLAlchemy default store
• scheduled_tasks/tasks.py — One-shot email delivery jobs now use jobstore="email" (in-memory) instead of the SQLAlchemy store

Why this works

• One-shot email delivery jobs don't need DB persistence — the EmailNotification table already tracks all email state (PENDING → SENDING → SENT/FAILED).
• Using an in-memory store for these ephemeral jobs removes the SQLAlchemy race condition entirely.
• Recurring jobs (process_email_queue, cleanup_email_otps) are unaffected and continue using the PostgreSQL job store.

Checklist

• I have commented my code, particularly in hard-to-understand areas
• I have added unit / integration tests that prove my fix is effective or that my feature works
• I have run all tests locally and they pass
• I have updated the documentation (if applicable)
• For any new secrets, I have updated the shared spreadsheet and the GitHub Secrets.

[marius-mather]

yep, looks like a nice way to do this