hexagon: add support for Q4_1 in MUL_MAT and MUL_MAT_ID (#23647)

* hex-mm: add support for Q4_1 matmul/matvec, hvx-only for now

* hmx-mm: add support for Q4_1

* hex-mm: use Q8_1 dynamic quantization to avoid having to compute sums in the vec_dot

* hexagon: fix repack scratch buffer overflow

* hex-mm: fix Q4_1 repack buffer sizing

* hexagon: flip the build order for mm and fa (seems to help LTO)

* hex-mm: add vec_dot 4x1s and minor HMX cleanup after adding Q4_1

* hex-mm: fix fp16 vec_dot fallback to 2x1 and another issue that could cause incorrect output

* hexagon: resurrect early-wake and add support for polling for op-batch completions

With Q4_1 ggml-hexagon now claims pretty much the entire graphs which gives the CPU more time to chilax.
This is a good thing! But it does add extra latency for the pure benchmark runs.
Early wakeup helps recover the latency a bit in the normals runs and op-batch polling is just for benchmarking.

---------

Co-authored-by: Todor Boinovski <todorb@qti.qualcomm.com>

ggml/src/ggml-hexagon/htp/main.c

@@ -853,6 +853,11 @@ static void htp_packet_callback(dspqueue_t queue, int error, void * context) {
         for (uint32_t i=0; i < n_ops; i++) {
             struct profile_data prof;
 
+            if (i == (n_ops-1)) {
+                // wake up the host before starting the last op
+                dspqueue_write_early_wakeup_noblock(queue, 0, 0);
+            }
+
             profile_start(ctx->profiler, &prof);
 
             proc_op_req(octx, tens, i, &ops[i]);
@@ -869,8 +874,6 @@ static void htp_packet_callback(dspqueue_t queue, int error, void * context) {
             }
         }
 
-        // dspqueue_write_early_wakeup_noblock(ctx->queue, 10, 0);
-
         struct htp_opbatch_rsp rsp;
         rsp.id        = req.id;
         rsp.status    = HTP_STATUS_OK;