Fixed an issue of getting 0-bitrate during reconnect

This issue was introduced by PR #5500 as it was not tested for reconnect.(Reconnect feature was broken when that PR was branched)

protocol/server/network.c

@@ -18,6 +18,7 @@ Includes
 
 #include <whist/core/whist.h>
 #include <whist/network/network.h>
+#include <whist/network/network_algorithm.h>
 #include <whist/input/input.h>
 #include <whist/logging/error_monitor.h>
 #include <whist/utils/aes.h>
@@ -381,6 +382,13 @@ int multithreaded_manage_client(void *opaque) {
 
         // When a new client has been connected, we want all threads to hold client active again
         reset_threads_holding_active_count(&state->client);
+
+        // Fill the network settings with default value, if we have a valid width and height
+        if (state->client_width > 0 && state->client_height > 0) {
+            udp_handle_network_settings(
+                state->client.udp_context.context,
+                get_default_network_settings(state->client_width, state->client_height));
+        }
         state->client.is_active = true;
     }
 

protocol/server/video.c

@@ -511,7 +511,7 @@ int32_t multithreaded_send_video(void* opaque) {
     while (!state->exiting) {
         update_client_active_status(&state->client, &assuming_client_active);
 
-        if (!assuming_client_active) {
+        if (!assuming_client_active || !state->client.is_active) {
             encoder_running = false;
             whist_sleep(1);
             continue;