[Metrics][Discover] Fix time range losing sync in metrics grid

[miloszmarcinkowski]

Summary

Closes #243167

The getDataRange function doesn't properly propagate selected values when users switch from Classic to ESQL view in Discover. We switched over to timeRange property provided by Discover plugin to solve this issue in the Metrics grid.

Before:

Screen.Recording.2025-11-18.at.17.23.35.mov

After:

Screen.Recording.2025-11-18.at.16.21.51.mov

How to test:

• Go to Discover and switch to ESQL mode
• Run TS metrics-*
• Switch to Classic mode
• Select absolute time range
• Switch to ESQL mode
• Select Last 15 minutes in the time range
• Verify if the correct time range is applied to Metrics grid.

[elasticmachine]

Pinging @elastic/obs-ux-infra_services-team (Team:obs-ux-infra_services)

[elasticmachine]

💚 Build Succeeded

• Buildkite Build
• Commit: a3b7938

Metrics [docs]

Async chunks

Total size of all lazy-loaded chunks that will be downloaded as the user navigates the app

id	before	after	diff
discover	1.2MB	1.2MB	-93.0B

[crespocarlos]

Thanks for this PR @miloszmarcinkowski.

I'd like to propose a different approach.

From what I gathered, what’s happening here is that the Metrics Experience profile misses the fetch emissions that happen while the profile or component is still loading, which makes it not call updateTimeRange. That may be amplified by the fact that the component is lazy-loaded (just a guess).

We may be able to solve this on our end by ensuring that input$ calls updateTimeRange once when it first get subscribed, and making sure this is utilized by both metrics and traces or any other component exposed in the plugin.

I'm not sure whether this can be solved on Discover side, but it may also be worth investigating.

wdyt?

[crespocarlos]

See comment above