Mojo function

generic_flare_mla_prefill_kv_cache_ragged

generic_flare_mla_prefill_kv_cache_ragged[collection_t: KVCollectionT, dtype: DType, //, softmax_type: DType, write_softmax_info: Bool, use_cascade_attention: Bool, mask_str: StringSlice[StaticConstantOrigin], score_mod_str: StringSlice[StaticConstantOrigin], target: StringSlice[StaticConstantOrigin], local_window_size: Int = -1](q: LayoutTensor[dtype, layout, origin, element_layout=element_layout, layout_int_type=layout_int_type, linear_idx_type=linear_idx_type, masked=masked, alignment=alignment], k: LayoutTensor[dtype, layout, origin, element_layout=element_layout, layout_int_type=layout_int_type, linear_idx_type=linear_idx_type, masked=masked, alignment=alignment], v: LayoutTensor[dtype, layout, origin, element_layout=element_layout, layout_int_type=layout_int_type, linear_idx_type=linear_idx_type, masked=masked, alignment=alignment], buffer_row_offsets: LayoutTensor[DType.uint32, layout, origin, element_layout=element_layout, layout_int_type=layout_int_type, linear_idx_type=linear_idx_type, masked=masked, alignment=alignment], cache_offsets: LayoutTensor[DType.uint32, layout, origin, element_layout=element_layout, layout_int_type=layout_int_type, linear_idx_type=linear_idx_type, masked=masked, alignment=alignment], input_row_offsets: LayoutTensor[DType.uint32, layout, origin, element_layout=element_layout, layout_int_type=layout_int_type, linear_idx_type=linear_idx_type, masked=masked, alignment=alignment], kv_collection: collection_t, layer_idx: UInt32, scale: Float32, output: LayoutTensor[dtype, layout, origin, element_layout=element_layout, layout_int_type=layout_int_type, linear_idx_type=linear_idx_type, masked=masked, alignment=alignment], softmax_info: LayoutTensor[softmax_type, layout, origin, element_layout=element_layout, layout_int_type=layout_int_type, linear_idx_type=linear_idx_type, masked=masked, alignment=alignment], context: DeviceContextPtr, prev_output: OptionalReg[LayoutTensor[dtype, Layout.row_major[3](), MutableAnyOrigin]] = None, prev_softmax_info: OptionalReg[LayoutTensor[softmax_type, Layout.row_major[3](), MutableAnyOrigin]] = None)