Decrease KOPS_SCHEDULER_QPS/BURST, KOPS_CONTROLLER_MANAGER_QPS/BURST to 300

Motivation / Background:
The ec2-master-scale-tests have been failing since September 19th, 2025 and possibly before due to PodStartUpLatency > 5s as per SLO.

Solution:
Historically, there were an experiment to increase scheduler throughput in 5k performance tests. The experiment has suggested
KOPS_SCHEDULER_QPS=300 and KOPS_SCHEDULER_BURST=300 to be ran at 300qps.  In addition, @wojtek stated that having a QPS at 300 would
hugely decrease pod-start-up-latency across the whole test.

The value of controller manager are modified to be uniformed with the the test-infra.

Signed-off-by: Ronald Ngounou <ronald.ngounou@yahoo.com>

Author: ronaldngounou

tests/e2e/scenarios/scalability/run-test.sh

@@ -39,10 +39,10 @@ if [[ -z "${ADMIN_ACCESS:-}" ]]; then
 fi
 echo "ADMIN_ACCESS=${ADMIN_ACCESS}"
 
-KOPS_SCHEDULER_QPS="${KOPS_SCHEDULER_QPS:-500}"
-KOPS_SCHEDULER_BURST="${KOPS_SCHEDULER_BURST:-500}"
-KOPS_CONTROLLER_MANAGER_QPS="${KOPS_CONTROLLER_MANAGER_QPS:-500}"
-KOPS_CONTROLLER_MANAGER_BURST="${KOPS_CONTROLLER_MANAGER_BURST:-500}"
+KOPS_SCHEDULER_QPS="${KOPS_SCHEDULER_QPS:-300}"
+KOPS_SCHEDULER_BURST="${KOPS_SCHEDULER_BURST:-300}"
+KOPS_CONTROLLER_MANAGER_QPS="${KOPS_CONTROLLER_MANAGER_QPS:-300}"
+KOPS_CONTROLLER_MANAGER_BURST="${KOPS_CONTROLLER_MANAGER_BURST:-300}"
 KOPS_APISERVER_MAX_REQUESTS_INFLIGHT="${KOPS_APISERVER_MAX_REQUESTS_INFLIGHT:-800}"
 ETCD_QUOTA_BACKEND_BYTES="${ETCD_QUOTA_BACKEND_BYTES:-8589934592}"
 echo "KOPS_SCHEDULER_QPS=${KOPS_SCHEDULER_QPS} KOPS_SCHEDULER_BURST=${KOPS_SCHEDULER_BURST}"