[Merge pending] add search ratio for limit forwarding search k size for agent & add disable_async_rerank option to use primitive sort rerank

[kpango]

Description:

This PR includes 2 feature.

1. Efficient Search-K forwarding for Agent using Query ratio paramerter. - Breaking: user should set 1 (100%) as default search query. - This PR requires minor update due to the breaking changes to proto payload.
2. Add non-async merging sort for low Top-K with low pod replica users which may faster to use lock and append and sort.

Related Issue:

Versions:

• Go Version: 1.20.2
• Docker Version: 20.10.8
• Kubernetes Version: 1.22.0
• NGT Version: 2.0.9

Checklist:

• [ ] I have read the CONTRIBUTING document and completed our CLA agreement.
• [ ] I have checked open Pull Requests for the similar feature or fixes?

Special notes for your reviewer:

[cloudflare-workers-and-pages[bot]]

Deploying vald with    Cloudflare Pages

Latest commit:	1cf0c47
Status:	✅  Deploy successful!
Preview URL:	https://6e418d0a.vald.pages.dev
Branch Preview URL:	https://feature-gateway-lb-add-searc.vald.pages.dev

View logs

[vdaas-ci]

[CHATOPS:HELP] ChatOps commands.

• :ok_woman: /approve - approve
• :love_letter: /changelog - replace the PR body by changelog details
• :bento: /format - format codes and add licenses
• :white_check_mark: /gen-test - generate test codes
• :label: /label - add labels
• :rewind: /rebase - rebase main
• :end: :two: :end: /label actions/e2e-deploy - run E2E deploy & integration test

[codecov[bot]]

Codecov Report

Attention: Patch coverage is 0% with 224 lines in your changes are missing coverage. Please review.

Project coverage is 29.43%. Comparing base (8bf4b82) to head (686f6f5). Report is 1 commits behind head on main.

:exclamation: Current head 686f6f5 differs from pull request most recent head 1cf0c47. Consider uploading reports for the commit 1cf0c47 to get more accurate results

Files	Patch %	Lines
pkg/gateway/lb/handler/grpc/handler.go	0.00%	220 Missing :warning:
hack/benchmark/internal/operation/search.go	0.00%	4 Missing :warning:

Additional details and impacted files

@@             Coverage Diff             @@
##             main    #1982       +/-   ##
===========================================
+ Coverage   18.89%   29.43%   +10.54%     
===========================================
  Files         562      365      -197     
  Lines       64595    34438    -30157     
===========================================
- Hits        12204    10138     -2066     
+ Misses      51597    23881    -27716     
+ Partials      794      419      -375     

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

[vankichi]

Could you please tell me the background and the benefits at the PR description?

[kpango]

@vankichi

Could you please tell me the background and the benefits at the PR description?

Due to the changes in the Search specification, the usage of the User's Client library will change, so a " Minor" equivalent release is required (Client backward incompatibility), and the Proto Index has been reorganized in this release.

[vdaas-ci]

[GEN TEST] Failed to generate tests.