server: also free the compute-graph scheduler on KV eviction

Under LLAMA_SLEEP_EVICT_KV a cold model still held the scheduler's worst-case
compute buffer (hundreds of MiB, e.g. ~700 MiB for gemma-26B) plus the resident
experts. release_device(evict_kv) now also frees the sched (sched.reset() +
sched_need_reserve), and restore_device() rebuilds it via sched_reserve(), so a
fully-evicted model holds essentially no VRAM (gemma cold: 7147 -> 51 MiB).
Trades a heavier re-warm (weights+KV H2D + sched reserve) for the freed VRAM;
weights-only mode is unchanged (fast switch, keeps KV+compute).

Assisted-by: Claude

src/llama-context.cpp

@@ -738,11 +738,15 @@ void llama_context::release_device(bool evict_kv) {
     if (evict_kv && memory && !kv_device_evicted) {
         memory->release_device_buffers();
         kv_device_evicted = true;
+        // also free the scheduler and its worst-case compute buffer (hundreds of MiB) so a cold
+        // model holds essentially no VRAM. Rebuilt lazily by sched_reserve() on restore.
+        sched.reset();
+        sched_need_reserve = true;
     }
 }
 
 void llama_context::restore_device() {
-    if (model.weights_resident() && !kv_device_evicted) {
+    if (model.weights_resident() && !kv_device_evicted && sched) {
         return;
     }
     model.restore_device_weights();
@@ -750,6 +754,10 @@ void llama_context::restore_device() {
         memory->restore_device_buffers();
         kv_device_evicted = false;
     }
+    // rebuild the scheduler + compute buffer if they were freed on release (evict_kv mode)
+    if (!sched) {
+        sched_reserve();
+    }
     // make sure all weight/KV uploads have completed before any compute reads them
     if (sched) {
         ggml_backend_sched_synchronize(sched.get());