Refactor scheduler lock implementation with heartbeat mechanism

- Add heartbeat-based lock renewal in scheduler_lock_heartbeat.py
- Update scheduler_lock.py with improved lock management
- Add comprehensive tests for scheduler lock functionality
- Update deployment and task documentation

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Docs/Tasks.md

@@ -1,50 +1,3 @@
-## [IMPORTANT] Database transactions lack explicit isolation
-
-**Where found**
-
-- `backend/app/repositories/session_repo.py:40-60` — multiple queries without `BEGIN TRANSACTION`
-- Similar pattern in multi-step operations across repositories
-
-**Why this is needed**
-
-Without explicit boundaries, concurrent requests can race: Thread A checks if exists → not found, Thread B checks same → not found, Thread A inserts → succeeds, Thread B inserts → duplicate error or silent overwrite.
-
-**Goal**
-
-Wrap all multi-step operations in explicit transactions with appropriate isolation level.
-
-**What to do**
-
-1. Use explicit `BEGIN IMMEDIATE` transaction:
-   ```python
-   await db.execute("BEGIN IMMEDIATE")
-   try:
-       await db.execute("INSERT INTO sessions ...")
-       await db.commit()
-   except Exception:
-       await db.rollback()
-       raise
-   ```
-
-2. Use `IMMEDIATE` mode to lock immediately for writes
-3. Document transaction boundaries clearly
-
-**Possible traps and issues**
-
-- Nested transactions (SAVEPOINTs) may be needed
-- Locks held too long cause contention
-- Deadlocks possible with concurrent writers
-
-**Docs changes needed**
-
-- Add section in `Docs/Backend-Development.md` § Database Transactions
-
-**Doc references**
-
-- `Docs/Backend-Development.md` (database design)
-
----
-
 ## [IMPORTANT] Scheduler lock race condition
 
 **Where found**