Remove forced PK_NewTasks index hint in _LockNextTask; add covering index for lock-next scan

[Copilot]

_LockNextTask hard-coded WITH (INDEX (PK_NewTasks)) forcing a clustered index scan for every dequeue. With a backlog of locked/invisible tasks this scans large numbers of ineligible rows on every poll — under high concurrency (N workers × M cores × ~20 polls/sec) this produces severe CPU amplification on Azure SQL.

Changes

• logic.sql — Remove WITH (INDEX (PK_NewTasks)) from the UPDATE TOP (1) in _LockNextTask, allowing the optimizer to choose the plan freely (consistent with _LockNextOrchestration, which has no forced hint):

  -- Before
  FROM NewTasks WITH (INDEX (PK_NewTasks))
  
  -- After
  FROM NewTasks

• schema-1.7.0.sql (new) — Add a covering index over the exact WHERE predicates used by _LockNextTask, giving the optimizer an efficient seek path to unlocked, visible tasks:

  CREATE NONCLUSTERED INDEX IX_NewTasks_LockNext
  ON NewTasks (TaskHub, LockExpiration, VisibleTime)
  INCLUDE (SequenceNumber, LockedBy, DequeueCount)

• common.props — Version bump 1.6.0 → 1.7.0 to trigger schema migration on upgrade and apply the new index to existing deployments.

Warning

Firewall rules blocked me from connecting to one or more addresses (expand for details)

I tried to connect to the following addresses, but was blocked by firewall rules:

• www.nuget.org
  • Triggering command: /usr/bin/dotnet dotnet build src/DurableTask.SqlServer/DurableTask.SqlServer.csproj (dns block)
  • Triggering command: /usr/bin/dotnet dotnet build src/DurableTask.SqlServer/DurableTask.SqlServer.csproj --no-restore (dns block)

If you need me to access, download, or install something from one of these locations, you can either:

• Configure Actions setup steps to set up my environment, which run before the firewall is enabled
• Add the appropriate URLs or hosts to the custom allowlist in this repository's Copilot coding agent settings (admins only)