Method build_attn_inp_kv_iswa

subprojects/llama.cpp/src/llama-graph.cpp:2101–2139

TODO: maybe separate the inner implementation into a separate function like with the non-sliding window equivalent once sliding-window hybrid caches are a thing.

//       like with the non-sliding window equivalent
//       once sliding-window hybrid caches are a thing.
llm_graph_input_attn_kv_iswa * llm_graph_context::build_attn_inp_kv_iswa() const {
    const auto * mctx_cur = static_cast<const llama_kv_cache_iswa_context *>(mctx);

    auto inp = std::make_unique<llm_graph_input_attn_kv_iswa>(hparams, cparams, mctx_cur);

    const auto n_stream = cparams.kv_unified ? 1 : ubatch.n_seqs_unq;

    {
        const auto n_kv = mctx_cur->get_base()->get_n_kv();

        inp->self_k_idxs = mctx_cur->get_base()->build_input_k_idxs(ctx0, ubatch);
        inp->self_v_idxs = mctx_cur->get_base()->build_input_v_idxs(ctx0, ubatch);

        inp->self_kq_mask = ggml_new_tensor_4d(ctx0, GGML_TYPE_F32, n_kv, n_tokens/n_stream, 1, n_stream);
        ggml_set_input(inp->self_kq_mask);
        ggml_set_name(inp->self_kq_mask, "self_kq_mask");

        inp->self_kq_mask_cnv = cparams.flash_attn ? ggml_cast(ctx0, inp->self_kq_mask, GGML_TYPE_F16) : inp->self_kq_mask;
        ggml_set_name(inp->self_kq_mask_cnv, "self_kq_mask_cnv");
    }

    {
        GGML_ASSERT(hparams.swa_type != LLAMA_SWA_TYPE_NONE && "Use llama_kv_cache for non-SWA");

        const auto n_kv = mctx_cur->get_swa()->get_n_kv();

        inp->self_k_idxs_swa = mctx_cur->get_swa()->build_input_k_idxs(ctx0, ubatch);
        inp->self_v_idxs_swa = mctx_cur->get_swa()->build_input_v_idxs(ctx0, ubatch);

        inp->self_kq_mask_swa = ggml_new_tensor_4d(ctx0, GGML_TYPE_F32, n_kv, n_tokens/n_stream, 1, n_stream);
        ggml_set_input(inp->self_kq_mask_swa);
        ggml_set_name(inp->self_kq_mask_swa, "self_kq_mask_swa");

        inp->self_kq_mask_swa_cnv = cparams.flash_attn ? ggml_cast(ctx0, inp->self_kq_mask_swa, GGML_TYPE_F16) : inp->self_kq_mask_swa;
        ggml_set_name(inp->self_kq_mask_swa_cnv, "self_kq_mask_swa_cnv");
    }

    return (llm_graph_input_attn_kv_iswa *) res->add_input(std::move(inp));
}

ggml_tensor * llm_graph_context::build_rs(
        ggml_tensor * s,