Initial extraction from isis: direct /dev/kfd GPU compute

AMD RDNA3 GPU driver via raw KFD ioctls. No ROCm, no OpenCL.

- Device discovery and topology queries
- VRAM/GTT/Userptr memory with shared address space
- PM4 compute queue and kernel dispatch
- Hand-written RDNA3 ASM kernels: matmul (3100 GFLOP/s), matvec, superlinear
- Tile-safe buffer padding for OOB protection
- Event-based interrupt wait (no CPU polling)

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

target/debug/.fingerprint/libc-6578899ebd06082d/build-script-build-script-build.json

@@ -0,0 +1 @@
+{"rustc":18276270781310494267,"features":"[\"default\", \"std\"]","declared_features":"[\"align\", \"const-extern-fn\", \"default\", \"extra_traits\", \"rustc-dep-of-std\", \"rustc-std-workspace-core\", \"std\", \"use_std\"]","target":5408242616063297496,"profile":1565149285177326037,"path":1233441601326685821,"deps":[],"local":[{"CheckDepInfo":{"dep_info":"debug/.fingerprint/libc-6578899ebd06082d/dep-build-script-build-script-build","checksum":false}}],"rustflags":[],"config":8247474407144887393,"compile_kind":0}