[Improvement] Liveness check and health check

### What would you like to be improved?

In an ideal environment, we should have 2 type of checks:
1. Liveness check: to determine if a container has started or not
2. Readiness check: to determine if the container is ready to accept network traffic.

A real example - if db has not been init yet while we deploy the service, the liveness check should pass while the readiness check should fail.

Currently Gravitino has the `api/version` endpoint that is more suitable for liveness check. Is there an endpoint for readiness check?

### How should we improve?

_No response_

Start with the existing api/version endpoint and trace how service startup and database initialization are exposed. Determine the readiness behavior needed for a container that has started but cannot yet accept network traffic, then document or test the endpoint behavior so liveness and readiness are distinguishable.

Written by the indexing model from the issue text.