Add tier_preference and node.role columns to _cat/shards API

[ramzeta]

This enhancement adds two new columns to the _cat/shards API to facilitate troubleshooting of ILM-related shard allocation issues:

• tier_preference: Shows the index.routing.allocation.include._tier_preference setting for the shard's index
• node.role: Displays node roles in abbreviated format (similar to _cat/nodes)

These columns help diagnose shards stuck during tier migrations (hot -> warm -> cold) by providing a consolidated view of tier preferences and node roles.

Closes #136895

[cla-checker-service]

💚 CLA has been signed

[pswaao88]

Hi @ramzeta, thanks for your contribution. However, I was already assigned to this issue (#136895) last week and I am currently working on it as discussed with the maintainer. I noticed this PR just now. Since I have been officially assigned, I would appreciate it if we could follow the assignment process.

[ramzeta]

Hi @pswaao88,

Thank you for bringing this to my attention. I apologize - I wasn't aware that you had been officially assigned to issue #136895 last week. I should have checked the assignment status before starting work on this.

I completely respect the assignment process and your work on this issue. I have a few options in mind:

1. Close this PR - If you prefer to continue with your implementation independently, I'm happy to close this PR to avoid duplicate efforts.

2. Collaborate - If you're open to it and the maintainers agree, perhaps we could collaborate on the solution. I'd be glad to contribute ideas or help review your approach.

3. Wait for maintainer guidance - We could wait for a maintainer like @DaveCTurner to weigh in on how to proceed.

I'm fine with any of these options. Please let me know what you prefer, and again, my apologies for the confusion.

Regarding the CLA - I have signed the Contributor Agreement, so that should be resolved once verified.

Best regards

[pswaao88]

Hi @ramzeta,

Thank you so much for the thoughtful message — I really appreciate it.

I actually just started looking into this issue recently and I’m currently running tests and trying to prepare a working implementation. That said, I truly don’t want to interrupt or duplicate any work you’ve already been doing.

I’m completely open to whatever approach the maintainers think is best.
Whether that means you continue with your PR, we collaborate, or I step back — I’m happy to follow their guidance.

Thanks again for reaching out so politely, and sorry for any overlap on my side as well.