server : add server_task_type field to server_task_result

This commit adds a server_task_type field to the server_task_result
struct. This field is used to identify the type of the server task.

The motivation for adding this is that it might allow us to avoid using
dynamic_cast's when checking the type of the server_task_result.
For example, this could then be replaced with checks like this:
```c++
        GGML_ASSERT(result.get() != nullptr);
        GGML_ASSERT(result.get()->get_server_task_type() == type);
```

examples/server/server.cpp

@@ -68,6 +68,7 @@ enum server_task_type {
     SERVER_TASK_TYPE_SLOT_RESTORE,
     SERVER_TASK_TYPE_SLOT_ERASE,
     SERVER_TASK_TYPE_SET_LORA,
+    SERVER_TASK_TYPE_NONE,
 };
 
 enum oaicompat_type {
@@ -480,6 +481,7 @@ struct result_timings {
 struct server_task_result {
     int id           = -1;
     int id_slot      = -1;
+    server_task_type type;
     virtual bool is_error() {
         // only used by server_task_result_error
         return false;
@@ -491,6 +493,7 @@ struct server_task_result {
     virtual int get_index() {
         return -1;
     }
+    virtual server_task_type get_server_task_type() = 0;
     virtual json to_json() = 0;
     virtual ~server_task_result() = default;
 };
@@ -794,6 +797,10 @@ struct server_task_result_cmpl_final : server_task_result {
 
         return ret;
     }
+
+    server_task_type get_server_task_type() {
+        return SERVER_TASK_TYPE_COMPLETION;
+    }
 };
 
 struct server_task_result_cmpl_partial : server_task_result {
@@ -962,6 +969,10 @@ struct server_task_result_cmpl_partial : server_task_result {
 
         return std::vector<json>({ret});
     }
+
+    server_task_type get_server_task_type() {
+        return SERVER_TASK_TYPE_NONE;
+    }
 };
 
 struct server_task_result_embd : server_task_result {
@@ -997,6 +1008,10 @@ struct server_task_result_embd : server_task_result {
             {"tokens_evaluated", n_tokens},
         };
     }
+
+    server_task_type get_server_task_type() {
+        return SERVER_TASK_TYPE_EMBEDDING;
+    }
 };
 
 struct server_task_result_rerank : server_task_result {
@@ -1016,6 +1031,10 @@ struct server_task_result_rerank : server_task_result {
             {"tokens_evaluated", n_tokens},
         };
     }
+
+    server_task_type get_server_task_type() {
+        return SERVER_TASK_TYPE_RERANK;
+    }
 };
 
 // this function maybe used outside of server_task_result_error
@@ -1071,6 +1090,10 @@ struct server_task_result_error : server_task_result {
     virtual json to_json() override {
         return format_error_response(err_msg, err_type);
     }
+
+    server_task_type get_server_task_type() {
+        return SERVER_TASK_TYPE_NONE;
+    }
 };
 
 struct server_task_result_metrics : server_task_result {
@@ -1127,6 +1150,10 @@ struct server_task_result_metrics : server_task_result {
             { "slots",                           slots_data },
         };
     }
+
+    server_task_type get_server_task_type() {
+        return SERVER_TASK_TYPE_METRICS;
+    }
 };
 
 struct server_task_result_slot_save_load : server_task_result {
@@ -1160,6 +1187,10 @@ struct server_task_result_slot_save_load : server_task_result {
             };
         }
     }
+
+    server_task_type get_server_task_type() {
+        return SERVER_TASK_TYPE_SLOT_SAVE;
+    }
 };
 
 struct server_task_result_slot_erase : server_task_result {
@@ -1171,12 +1202,20 @@ struct server_task_result_slot_erase : server_task_result {
             { "n_erased", n_erased },
         };
     }
+
+    server_task_type get_server_task_type() {
+        return SERVER_TASK_TYPE_SLOT_ERASE;
+    }
 };
 
 struct server_task_result_apply_lora : server_task_result {
     virtual json to_json() override {
         return json {{ "success", true }};
     }
+
+    server_task_type get_server_task_type() {
+        return SERVER_TASK_TYPE_NONE;
+    }
 };
 
 struct server_slot {
@@ -2751,6 +2790,11 @@ struct server_context {
                     res->id = task.id;
                     queue_results.send(std::move(res));
                 } break;
+            case SERVER_TASK_TYPE_NONE:
+                {
+                    // do nothing
+                    GGML_ASSERT(false && "Invalid task.type (SERVER_TASK_TYPE_NONE)\n");
+                } break;
         }
     }
 
@@ -3693,12 +3737,8 @@ int main(int argc, char ** argv) {
             res_error(res, result->to_json());
             return;
         }
-
+        GGML_ASSERT(result.get() != nullptr);
-        if (type == SERVER_TASK_TYPE_SLOT_SAVE) {
+        GGML_ASSERT(result.get()->get_server_task_type() == type);
-            GGML_ASSERT(dynamic_cast<server_task_result_slot_save_load*>(result.get()) != nullptr);
-        } else if (type == SERVER_TASK_TYPE_SLOT_ERASE) {
-            GGML_ASSERT(dynamic_cast<server_task_result_slot_erase*>(result.get()) != nullptr);
-        }
         res_ok(res, result->to_json());
     };