refactor: remove bf16 gemm and swiglu kernels and rebuild csrc benchmarks

- delete csrc/kernels/gemm.cu and swiglu.cu and drop their CMake and setup.py registration
- remove the ops wrappers plus backend/linear.py and backend/swiglu.py so Linear and MLP call F.linear directly
- drop the four gemm and swiglu kernel test files and prune the stale cuda_kernels.md sections
- add csrc/bench benchmarks for the remaining kernels: attention decode prefill paged decode paged prefill versus single-launch SDPA references, rotary versus the torch fallback, fp8 quantize and mm_fp8 versus torch baselines
- attention, rotary_emb, and fp8_ops kernels are unchanged

csrc/CMakeLists.txt

@@ -61,8 +61,6 @@ set(KERNEL_NAMES
     attn_prefill
     attn_paged_decode
     attn_paged_prefill
-    bf16_gemm
-    bf16_swiglu
     rotary_emb
 )
 set(KERNEL_SRCS
@@ -70,8 +68,6 @@ set(KERNEL_SRCS
     attention/prefill.cu
     attention/paged_decode.cu
     attention/paged_prefill.cu
-    gemm.cu
-    swiglu.cu
     rotary_emb.cu
 )