perf: rebuild decode gemm dispatch around shape-driven tile configs

- split-K removed entirely: tiled kernel walks K in one pass, no partials/semas workspace, no memset, single launch per call
- skinny GEMM (M<=8) dispatch table replaces the hand-written switch
- shape-driven four-family table replaces plan_gemm: wide-N (n>=4096) default {16,64,64,3,128} with BM=32 at M>16; narrow-N deep-K rings {16,32,256,2,64} while the grid fits one wave, {16,32,128,2,64} past it
- narrow-N is K-serial: widening the grid measurably does nothing (BN 64->32 ties, doubled m_tiles tie, kv at 4 blocks ties q/o at 24); deeper K chunks win until 72KB smem forces one CTA per SM and past one wave the 2-wave quantization loses to BK=128
- launch-check macros in common/launch.cuh; smem opt-in for the 72KB/60KB rings
- rename kernels/bf16_*.cu to gemm.cu/swiglu.cu; module names unchanged
- Python gate: lm_head (N>32768) falls back to cuBLAS, band narrows to M<=32
- drop the stale per-op benchmark narratives; fold the live numbers into cuda_kernels.md

Benchmark: NVIDIA L20 (sm_89, 92 SMs), CUDA 12.8, bf16, L2-thrash weight rotation, per-call medians at M=16: q/o 9.5us, kv 8.6us, gate/up 33.3us, down 33.7us (down -29% vs prior default). End-to-end 1B decode (gen 128, 3 trials, tokens/s vs cuBLAS): B=1 260 vs 252, B=8 1660 vs 1446, B=16 2464 vs 2437, B=32 3620 vs 3690. Prior split-K dispatch measured B=16 2243 / B=32 3393.

tests/extension/test_swiglu_dispatch.py

@@ -86,10 +86,9 @@ def test_mode_one_forces_supported_shape(monkeypatch):
 
 
 @skip_no_swiglu
-def test_auto_uses_unfused_chain_until_shape_is_qualified(monkeypatch):
-    # The fusion table is empty, so auto keeps the unfused linear-backend
-    # chain. The linear backend may still dispatch its own GEMV for M=4,
-    # hence the relaxed tolerance versus the pure-torch reference.
+def test_auto_uses_fused_chain_for_decode_batches(monkeypatch):
+    # Auto adopts the fused primitive for the decode band; numerics match
+    # the unfused linear-backend chain within BF16 accumulation-order noise.
     monkeypatch.setenv("ASTRAI_SWIGLU", "auto")
     x = torch.randn(4, 1536, device="cuda", dtype=torch.bfloat16)
     up_weight = torch.randn(6912, 1536, device="cuda", dtype=torch.bfloat16)