Flatten remaining user-guide subtrees (partial fix for #1044 #1051)

[OVI3D0]

Description

PR C2 in the flatten stack. C1 (#12788) landed and this continues the same esrally-inspired approach: collapse deep _benchmark/user-guide/* subtrees into top-level pages under _benchmark/.

This PR now covers the remaining user-guide subtrees in a single change (previously planned as separate C2 and C3 PRs). Combined, they are still small: 16 files touched, +31/-102 lines.

Files moved

From _benchmark/user-guide/optimizing-benchmarks/ → _benchmark/:

• distributed-load.md
• expand-data-corpus.md
• performance-testing-best-practices.md
• randomizing-queries.md
• target-throughput.md

From _benchmark/user-guide/working-with-workloads/ → _benchmark/:

• running-workloads.md
• creating-custom-workloads.md
• finetune-workloads.md
• contributing-workloads.md

Each moved file has a redirect_from entry for its just-vacated subtree path. Files that already had redirect_from entries (target-throughput.md, the working-with-workloads pages) preserve them and pick up the new one. creating-custom-workloads.md also drops a redundant self-referential redirect_from entry that pointed at its new canonical URL.

Files deleted

• _benchmark/user-guide/optimizing-benchmarks/index.md (subtree landing page)
• _benchmark/user-guide/understanding-results/index.md (subtree landing page — this directory was already empty after #12394 unified telemetry and #12734 consolidated summary-report)
• _benchmark/user-guide/working-with-workloads/index.md (subtree landing page)

Cross-references updated

Three cards in _benchmark/user-guide/index.md pointed at the just-deleted landing pages. They now point at canonical successor pages:

• "Understanding your results" → reference/summary-report/
• "Optimizing benchmarks" → performance-testing-best-practices/
• "Run and customize your workload" → running-workloads/

Also cleaned up a few lingering absolute links in anatomy-of-a-workload.md, choosing-a-workload.md, and _benchmark/index.md that pointed at the pre-C2 paths. The redirect_from entries would have kept them working, but the new paths are cleaner.

Issues Resolved

Partial fix for:

• Flatten + Reorganize Documentation Structure opensearch-benchmark#1044
• [Documentation] Flatten OSB's Documentation Structure opensearch-benchmark#1051

After this PR, the _benchmark/user-guide/ subtree contains only concepts.md and install-and-configure/ — candidates for a follow-up PR if maintainers want to flatten further.

Version

all

Validation

• Local Jekyll build with JEKYLL_LINK_CHECKER=internal completed cleanly (exit 0, zero errors, zero warnings).
• All old-path URLs return 200 locally as redirect_from stubs that meta-refresh to the canonical new URLs.
• No lingering intra-doc references to the old paths (grep verified — the only remaining hits are the intentional redirect_from entries).

Checklist

• By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license and subject to the Developers Certificate of Origin.

[github-actions]

Thank you for submitting your PR. The PR states are In progress (or Draft) -> Tech review -> Doc review -> Merged.

Before you submit your PR for doc review, make sure the content is technically accurate. If you need help finding a tech reviewer, tag a maintainer.

When you're ready for doc review, tag the assignee of this PR. The doc reviewer may push edits to the PR directly or leave comments and editorial suggestions for you to address (let us know in a comment if you have a preference).

[OVI3D0]

@kolchfa-aws ready for review!

cc @rishabh6788

[kolchfa-aws]

@OVI3D0 When you move pages, you need to add redirects from any page that you deleted to an existing page. If an index page is deleted, we need to add a redirect to one of the surviving pages. I added those as part of the doc review. Also, reordered the pages in the top-level nav as follows:

OpenSearch Benchmark
Quickstart
User guide (section)
Anatomy of a workload
Choosing a workload
Common operations
Running a workload
Creating custom workloads
Fine-tuning custom workloads
Sharing custom workloads
Running distributed loads
Expanding a workload's data corpus
Target throughput
Randomizing queries
Performance testing best practices
Reference (section)
Additional features (section)
Vector search
Glossary
Version history
Migration assistance
FAQs

For now, I left vector search on the top level, but it's a workload type and it belongs with the other workload types. So I would restructure as follows, keeping in mind that the current Anatomy of a workload and Workload reference have some overlap:

Choosing a workload // just keep the task-based info here that helps the user decide what workload type to choose
Workload types  // move workload summary sections here from Choosing a workload              
  ├─ Vector search             
Anatomy of a workload
  ├─ indices
  ├─ corpora
  ├─ operations                
  ├─ test procedures
  └─ workload parameters