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>

topology.yaml

@@ -65,6 +65,7 @@ cube:
       pe_dma:       { kind: pe_dma,       impl: pe_dma_v1,       attrs: { rd_engines: 1, wr_engines: 1 } }
       pe_gemm:      { kind: pe_gemm,      impl: pe_gemm_v1,      attrs: { overhead_ns: 0.0, shared_resource: accel_slot, peak_tflops_f16: 8.0 } }
       pe_math:      { kind: pe_math,      impl: pe_math_v1,      attrs: { overhead_ns: 0.0, shared_resource: accel_slot } }
+      pe_mmu:       { kind: pe_mmu,       impl: pe_mmu_v1,       attrs: { tlb_overhead_ns: 0.5, page_size: 4096 } }
       pe_tcm:       { kind: pe_tcm,       impl: pe_tcm_v1,       attrs:
       { size_mb: 16 } }
     links: