Add SchedulerV2 (pe_accel), DPPolicy overrides, and new benchmarks

- Add cycle-accurate PE accelerator scheduler (SchedulerV2) with tiled
  GEMM/Math pipelines (DMA_IN → GEMM → MATH → DMA_WB)
- Add DPPolicy num_pes/num_cubes/num_sips overrides for single-PE testing
- Support tuple target_pe for targeting specific PE subsets
- Add gemm_single_pe and gpt3_qkv benchmarks
- Switch default topology to pe_scheduler_v2

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

src/kernbench/runtime_api/kernel.py

@@ -89,7 +89,7 @@ class KernelLaunchMsg:
     kernel_ref: KernelRef
     args: tuple[KernelArg, ...]
     target_cubes: tuple[int, ...] | Literal["all"] = "all"
-    target_pe: int | Literal["all"] = "all"
+    target_pe: int | tuple[int, ...] | Literal["all"] = "all"
     msg_type: Literal["kernel_launch"] = "kernel_launch"