feat: add independent native-precision sparse SDPA kernel path
Wire up the dormant SPARSESDPAFwdMainloop into a new SparseSDPAConfig with
ElementQ-aware (bf16/fp16) MMA, plus launchers, a wrapper, declarations, and a
sage_sparse_sdpa pybind. Native-precision path applies the softmax scale directly
and carries no INT8 dequant baggage, unlike the shared sparse-SAGE path.