feat: Bot instance upgrade status dashboard

[nicholasmarais1158]

Summary

This change adds a dashboard to the bot instances list. It displays when there is no selected instance item. A breakdown of instance counts by upgrade status gives a high-level overview, and allows easy filtering to show instances for each status by clicking on an item on the dashboard. The metrics data is periodically auto-refetched based on a value (refresh_after_seconds) from the API (or every 1 min, as a fallback).

Updates: #55926
Depends on: #59896
Changelog: Added a dashboard to visualize bot instances by their upgrade status

Changes

• Added dashboard component to right-hand side of bot instance page
• Aligned the version compatibility logic with the backend (versions newer than the cluster are consisted "unsupported")
• Fixed a few stories that were no longer mocking the correct versioned API

Demo

Screen.Recording.2025-10-07.at.17.12.10.mov

Reviewer notes

Here's a script to insert a bunch of bot instances to make testing easier; bot_instances.sql

$ sqlite3 <your cluster's data dir>/backend/sqlite.db
sqlite> .read bot_instances.sql

Be sure to restart your cluster afterwards - the cache will not be notified of the changes.

[nicholasmarais1158]

Rebased on top of master after #59896 was merged.

[avatus]

Suggested change

	{error ? (
	<Alert kind="danger" m={3}>
	{error.message}
	</Alert>
	) : undefined}
	{error && (
	<Alert kind="danger" m={3}>
	{error.message}
	</Alert>
	)}

[nicholasmarais1158]

This is a style I've used throughout the frontend for bots. It's not the first challenge I've had, so I'll replace them in a separate PR dedicated to just that.

[bl-nero]

There may be a (perhaps fututre) opportunity to unify and reuse some code between this one and UsageSummary from our enterprise code.

[nicholasmarais1158]

Where do I find the Usage Dashboard in the product?

[bl-nero]

I think it appears in the dashboard mode, when you launch it as a cloud server. There were some additional prerequisites, but I frankly don't remember. It's probably easiest to just set up a cloud tenant.

[zmb3]

Haven't looked at the code, but a couple thoughts from the demo video:

1. I'd consider changing the "Upgrade Status" header. To me that implies checking the status of a rollout and watching things update. This is more like "version status" (though that's a terrible name, please don't use it.
2. I also wonder if we should change the "Currently Active" header depending on which group is selected on the right. When I saw you select "Patch Available" it was a little confusing to see "no active instances" with helper text explaining that instances expire after not heartbeating. The list is empty not because there are no active instances, but because there are no active instances with a patch release available.

[backport-bot-workflows]

@nicholasmarais1158 See the table below for backport results.

Branch	Result
branch/v17	Failed
branch/v18	Failed