docs: note inline data source picker for AIOps and data drift (#6879)

[florent-leborgne]

Summary

Closes #6879. Kibana #266672 removed the intermediate Select Data View page for AIOps tools and Data Visualizer pages, replacing it with an inline data view/Discover session picker in the page header. This PR brings the docs in line with that change.

Screenshots (8) retaken and replaced to show the current inline picker UI:

• explore-analyze/machine-learning/machine-learning-in-kibana.md: Data Visualizer, Data drift
• explore-analyze/machine-learning/machine-learning-in-kibana/xpack-ml-aiops.md: Log rate analysis, Log pattern analysis, Change point detection, Selected change points
• explore-analyze/machine-learning/anomaly-detection/ml-getting-started.md: two Data Visualizer field-statistics screenshots

Wording fixes:

• ml-getting-started.md told readers to "Click Select data view" — that button no longer exists. Reworded to describe the outcome ("Select the kibana_sample_data_logs data source") instead of naming a specific control, so the instruction stays accurate for both the pre- and post-9.5 UI without needing version scoping.
• Bolded page/feature names on first mention (Data drift, AIOps Labs, Log rate analysis, Log pattern analysis, Change point detection) for consistency with the rest of the corpus and with the style guide's "bold names shown in the UI" rule — verified against the retaken screenshots that each is a literal page header.
• Reframed the procedural passages on the two overview pages (machine-learning-in-kibana.md, xpack-ml-aiops.md) from click-by-click instructions into descriptive, conceptual voice, better matching the overview content type these pages are meant to be. No information was dropped.
• Added the missing frontmatter description field to all three pages.
• Fixed a batch of pre-existing style issues (word choice, voice/tone, grammar, a stale alt text) on the two overview pages while in there.

Lifecycle fix (Data drift): while auditing the applies_to/preview markers on these pages, found that the unscoped "technical preview" warning on Data drift was stale. Kibana issue #245828 and PR #252764 (v9.4.0) confirm the preview badge was removed from the product because the feature is "no longer in tech preview," and the ML team confirmed it was marked GA in stack 9.4. Replaced the warning with a section-level applies_to tag (stack: ga 9.4+, preview 9.0-9.3, serverless: ga) that preserves the preview state for the still-supported 9.0-9.3 range per the cumulative docs model, matching the pattern already used for Change point detection on the sibling AIOps Labs page. Also verified the two other lifecycle markers on these pages (the ga 9.4 Zoom in/out date-picker tip, and Change point detection's preview 9.0-9.4 → ga 9.5+ tag) against Kibana source/PR history — both are accurate, no change needed.

Why this closes the issue

The issue's core ask is that the docs accurately reflect the product after the intermediate selection page was removed. I audited the full docs-content corpus (not just these two pages) for the specific things the removal could have made stale:

• No remaining references to the removed Select Data View page, breadcrumb level, or the indexDataVisualizerPage deep-link, other than the one in ml-getting-started.md, now fixed.
• No "Index data visualizer (ES|QL)" variant is documented anywhere, so there's nothing to rename there.
• The existing "you can select the data view or saved Discover session" phrasing on both overview pages never named a specific selection mechanism (button vs. inline picker), so it was already accurate for both the old and new UI and didn't need a version-scoped rewrite.

I deliberately did not add a version-scoped note stating users can now switch data views "without leaving the page" (an earlier draft of this PR did). A picker sitting in the page header is self-evidently reusable — there's no discoverability gap for a reader to get stuck on, and the only thing such a note would really convey is a comparison to a UI that most readers never saw, which is changelog framing rather than evergreen reference content.

Not in scope (flagged, not fixed)

• xpack-ml-aiops.md structurally bundles three distinct tools (Log rate analysis, Log pattern analysis, Change point detection) in one overview page, each with enough procedural depth to function as its own how-to. Splitting these into linked pages would be a cleaner fit for the overview content type, but it's real IA work (new pages, redirects, cross-links) that deserves its own PR rather than being folded into this one.
• ml-getting-started.md has a larger set of pre-existing style issues (recurring "see"/"choose" word choice, gerund-led sentences, directional language) unrelated to this change — not touched here.

Generative AI disclosure

1. Did you use a generative AI (GenAI) tool to assist in creating this contribution?

• Yes
• No

Tool(s) and model(s) used: Claude Code (Claude Sonnet 5), with screenshots retaken manually by the author.

Resolves

Closes #6879

[github-actions]

Elastic Docs AI PR menu

Check the box to run an AI review for this pull request.

• Review docs changes (docs-review). Status: not started.

Powered by GitHub Agentic Workflows and docs-actions. For more information, reach out to the docs team.

[github-actions]

🔍 Preview links for changed docs

• explore-analyze/machine-learning/anomaly-detection/ml-getting-started.md
• explore-analyze/machine-learning/machine-learning-in-kibana.md
• explore-analyze/machine-learning/machine-learning-in-kibana/xpack-ml-aiops.md

[github-actions]

Elastic Docs Style Checker (Vale)

Summary: 1 warning found

⚠️ Warnings (1): Fix when the suggestion improves clarity or correctness.

File	Line	Rule	Message
explore-analyze/machine-learning/machine-learning-in-kibana.md	17	Elastic.Spelling	'data's' is a possible misspelling.

---

The Vale linter checks documentation changes against the Elastic Docs style guide. To use Vale locally or report issues, refer to Elastic style guide for Vale.

[nastasha-solomon]

This should be an applies_to section-level tag, no?