ADR-0001 Rev 2: 51-bit PhysAddr layout with concrete sub-unit tables

Remove rack_id (4 bits), rename sip_seg→die_id, shift fields to enable
42-bit local_offset (4 TB per die). Define PE_LOCAL/MCPU_LOCAL/CUBE_SRAM
sub-unit tables for AHBM dies and IOCPU sub-unit table for IOCHIPLET
dies (1 TB window). Supersedes ADR-0031.

Also fixes latent VA/PA confusion in pe_dma pipeline DMA path where
virtual addresses were decoded as physical addresses without MMU
translation — previously masked by coincidental bit-position alignment.

529 passed (+6 recovered), 10 pre-existing failures unchanged.

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

tests/test_tensor_free.py

@@ -55,7 +55,7 @@ def _make_ctx():
 
 def test_allocator_free_hbm_reclaims_space():
     """free_hbm returns HBM space; subsequent alloc can reuse it."""
-    a = PEMemAllocator(rack_id=0, sip_id=0, cube_id=0, pe_id=0, cfg=_CFG)
+    a = PEMemAllocator(sip_id=0, die_id=0, pe_id=0, cfg=_CFG)
     pa1 = a.alloc_hbm(4096)
     used_after_alloc = a.hbm_used
     a.free_hbm(pa1, 4096)
@@ -66,7 +66,7 @@ def test_allocator_free_hbm_reclaims_space():
 
 def test_allocator_free_tcm_reclaims_space():
     """free_tcm returns TCM space."""
-    a = PEMemAllocator(rack_id=0, sip_id=0, cube_id=0, pe_id=0, cfg=_CFG)
+    a = PEMemAllocator(sip_id=0, die_id=0, pe_id=0, cfg=_CFG)
     pa1 = a.alloc_tcm(256)
     used_after_alloc = a.tcm_used
     a.free_tcm(pa1, 256)