B1: mixed FP8/BF16 decode FMHA integration

- New: fmha_mixed_fp8_decode.cuh (Blackwell FP8 tensor-core FMHA kernel)
- New: fmha_mixed_fp8_capi.cu (C ABI launcher)
- New: fmha_mixed_fp8_op.py (Python ctypes/nvcc bridge)
- New: fp8_attention_io.cu (Q quantize + mixed KV gather kernels)
- New: fmha_umma_desc.cuh additions (f8f6f4 UMMA + idesc helpers)
- Modified: production.py (dsv4_attention_mixed_fp8_decode API)
- Modified: single_shot_inference.py (B1 gather + FMHA path)
- Modified: __init__.py (export mixed FP8 API)
- New: docs/B1_MIXED_FP8_FMHA.md, FINAL_STRETCH.md

noPE KV stays FP8_E4M3 + per-row scale, RoPE stays BF16.
No global FP8->BF16 KV staging before FMHA.
Decode-only (T==1), specialized HD=512/NOPE=448/ROPE=64.
CUDA compile/runtime validation pending on B200.

docs/B1_MIXED_FP8_FMHA.md

@@ -0,0 +1,55 @@
+# B1 Mixed FP8/BF16 FMHA first pass
+
+Implemented a decode-only DeepSeek-V4 attention path that keeps the cache in the paper/native storage format:
+
+- noPE KV: FP8_E4M3 bytes plus per-row FP32 scale
+- RoPE KV: BF16
+- Q noPE: quantized BF16 -> FP8_E4M3 immediately before FMHA
+- Q RoPE: BF16
+
+The live `forward_attention` path now gathers compressed rows and the SWA tail into mixed buffers and calls `dsv4_attention_mixed_fp8_decode`; it no longer dequantizes noPE KV into `gather_buf` before attention.
+
+## New files
+
+- `dsv4/kernels/cuda/fp8_attention_io.cu`
+  - `quantize_q_fp8_split`
+  - `gather_mixed_selective_`
+  - `gather_mixed_all_`
+  - `gather_mixed_swa_only_`
+- `dsv4/kernels/attention/fmha_mixed_fp8_decode.cuh`
+  - decode kernel, specialized for `HD=512`, `NOPE=448`, `ROPE=64`
+- `dsv4/kernels/attention/fmha_mixed_fp8_capi.cu`
+  - C ABI launcher
+- `dsv4/kernels/attention/fmha_mixed_fp8_op.py`
+  - Python ctypes/nvcc bridge
+
+## Modified files
+
+- `dsv4/kernels/attention/fmha_umma_desc.cuh`
+  - added `.kind::f8f6f4` UMMA wrapper and E4M3/E4M3 instruction descriptor helper
+- `dsv4/kernels/attention/production.py`
+  - added `dsv4_attention_mixed_fp8_decode`
+- `dsv4/kernels/attention/__init__.py`
+  - exported mixed FP8 API
+- `single_shot_inference.py`
+  - added mixed gather buffers/methods to `KVCache`
+  - changed step 5 gather to preserve FP8 noPE globally
+  - changed step 6 FMHA to call the mixed FP8 decode path
+
+## Intentional first-pass limits
+
+- Decode only (`T == 1`). The launcher hard-errors for prefill.
+- Specialized to DeepSeek-V4 attention dimensions (`512/448/64`).
+- noPE QK uses Blackwell FP8 tensor cores; RoPE QK and PV use BF16 tensor cores.
+- noPE V is dequantized only inside shared memory immediately before the PV BF16 tensor-core multiply. There is no global BF16 KV staging.
+
+## Validation status
+
+The sandbox used to make this patch does not have `nvcc`, so CUDA compilation/runtime validation was not possible here. Python syntax was checked with:
+
+```bash
+python3 -m py_compile single_shot_inference.py \
+  dsv4/kernels/attention/production.py \
+  dsv4/kernels/attention/fmha_mixed_fp8_op.py
+```
+