docs: update embedding benchmarks (3.9.6)

[github-actions]

Automated embedding benchmark update for 3.9.6 from workflow run #786.

[greptile-apps]

Greptile Summary

Automated benchmark update appending v3.9.6 results (1500 symbols, structured strategy, dated 2026-04-30) to the comparison table and the latest-results section. All Hit@N percentages, miss counts, and timing values match the raw JSON counts exactly — no data errors were found.

Confidence Score: 5/5

Safe to merge — documentation-only change with internally consistent data.

Single generated markdown file with no logic changes. All benchmark percentages (Hit@1/3/5/10), miss counts, and timing figures were verified against the raw JSON counts and are correct to one decimal place.

No files require special attention.

Important Files Changed

Filename	Overview
generated/benchmarks/EMBEDDING-BENCHMARKS.md	Appends v3.9.6 embedding benchmark rows; all percentages and miss counts verified correct against raw hit-count JSON.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A[Workflow run 786 - v3.9.6 benchmark job] --> B[Embed 1500 symbols with each model]
    B --> D[minilm - Hit@1 67.4%]
    B --> E[jina-small - Hit@1 75.8%]
    B --> F[jina-base - Hit@1 72.1%]
    B --> G[nomic - Hit@1 80.1%]
    B --> H[nomic-v1.5 - Hit@1 78.4%]
    D & E & F & G & H --> I[EMBEDDING-BENCHMARKS.md updated by github-actions bot]

Loading

_{Reviews (1): Last reviewed commit: "docs: update embedding benchmarks (3.9.6..." | Re-trigger Greptile}