Indices top_queries-* are permanently in YELLOW state

[lukeskywlkr]

Describe the bug
After upgrading to OpenSearch v2.19.1, indices matching the pattern top_queries-* remain permanently in a YELLOW state on single-node clusters. This is due to the presence of unassigned replica shards, which cannot be allocated in a single-node setup. As a result, the overall OpenSearch cluster health is reported as YELLOW, even though the system is otherwise functioning correctly.

To Reproduce
Steps to reproduce the behavior:

1. Install or upgrade to OpenSearch v2.19.1.
2. Deploy a single-node cluster.
3. Start the cluster and wait for it to initialize.
4. Observe the cluster health and note that indices named top_queries-* are in YELLOW state due to unassigned replicas.
5. Observe that the entire cluster reports YELLOW health status.

Expected behavior
In a single-node cluster, system indices such as top_queries-* should be created with number_of_replicas: 0, or the system should automatically handle this condition to avoid unnecessary YELLOW health status. The cluster should report GREEN health unless actual issues exist.

Screenshots
N/A

Desktop (please complete the following information):
N/A

Kubernetes (please complete the following information):

• Kubernetes version: 1.32.2
• Module Logging Version: v5.1.0

Additional context
This issue is caused by the new top_query feature in v2.19.1. It was fixed in PR #254, which adjusts the settings to prevent replica shards from being assigned where they cannot be allocated. The issue that tracked this bug is #256.