HDDS-15637. Container Replica Debugger Tool improvements and minor bug fixes

[sreejasahithi]

What changes were proposed in this pull request?

Following improvements and minor bug fixes need to be done for the tool:

1. Proper Database path validation
   currently users could pass a non-existent DB file path and fail later with less clear errors and with an empty db file crated in the current working directory.

2. Skipping invalid log files and make log filename parsing more robust using regex
   When a file is skipped because it is not a container log, the process should not get stuck waiting on it. We should do latch.countDown() in such cases to prevent the process to get stuck.
   Ensure that only valid container log files are processed and unrelated files present in the directory are ignored.

3. Health filter query fixes
   simplify and improve Over/under-replicated and unhealthy container sql queries.
   Rewrite SQL to count replicas correctly:
   Under-replicated: healthy replicas (not UNHEALTHY/DELETED) < replication factor
   Over-replicated: total active replicas > replication factor, and healthy replicas >= replication factor
   Unhealthy: all active replicas are UNHEALTHY

4. Add per-state replica summary to container analysis
   Enhance the info sub-command to display a summary of replica states after container analysis.
   Include counts for states such as CLOSED, QUASI_CLOSED, CLOSING, OPEN, DELETED, and UNHEALTHY to provide a quick overview of replica distribution.

5. Add result limiting support to duplicate-open sub-command
   Introduce options to limit the number of results returned by the duplicate-open sub-command.
   Provide an option to display all matching results when required.

What is the link to the Apache JIRA

HDDS-15637

How was this patch tested?

Green CI : https://github.com/sreejasahithi/ozone/actions/runs/27936681974

[sreejasahithi]

@ashishkumar50 Could you please review this PR.