Add virtual memory support: PE_MMU, VA allocator, fabric MmuMapMsg

Implement VA/MMU layer (ADR-0011 Phase 1) enabling Triton kernels to use
contiguous virtual addresses on sharded tensors.

Key changes:
- PE_MMU component: hybrid inbox (MmuMapMsg) + sync translate() for PE_DMA
- VirtualAllocator + PEMemAllocator: free-list with coalescing
- MmuMapMsg/MmuUnmapMsg fabric path with SIP-level routing
- DPPolicy-based mapping: replicate=local, sharded=broadcast
- Tensor lifecycle: del + weakref cleanup, context manager
- Rename: TensorHandle.pa→addr, DmaReadCmd.src_pa→src_addr, ctx→torch

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

src/kernbench/topology/builder.py

@@ -22,6 +22,7 @@ _PE_COMP_OFFSETS = {
     "pe_dma": (0.0, -0.15),
     "pe_gemm": (0.0, 0.0),
     "pe_math": (0.0, 0.15),
+    "pe_mmu": (0.15, -0.15),
     "pe_tcm": (0.3, 0.0),
 }
 
@@ -495,6 +496,15 @@ def _instantiate_cube(
                 kind="pe_response",
             ))
 
+            # noc → PE_MMU (MMU mapping install)
+            pe_mmu_id = f"{pp}.pe_mmu"
+            if pe_mmu_id in nodes:
+                edges.append(Edge(
+                    src=f"{cp}.noc", dst=pe_mmu_id,
+                    distance_mm=clinks.get("noc_to_pe_mmu_mm", 0.0),
+                    kind="command",
+                ))
+
             pe_idx += 1
 
     # ── xbar_top/bot → HBM slices ──
@@ -1073,6 +1083,7 @@ def _build_pe_view(spec: dict) -> ViewGraph:
         "pe_dma": (7.0, 1.5),
         "pe_gemm": (7.0, 4.0),
         "pe_math": (7.0, 6.5),
+        "pe_mmu": (4.0, 1.5),
         "pe_tcm": (10.0, 4.0),
     }