benchmark: integrate pprof analysis into reports

[avinxshKD]

Depends on #1285. This change is based on StLeoX:feat/router-benchmark

What type of PR is this?

/kind feature

What this PR does / why we need it:

Integrates pprof analysis into ResultReporter.build_report().

The reporter parses collected CPU and heap profiles, resolves samples to functions, and adds the hottest scheduler/plugin functions to each configuration's pprof_analysis report section. Individual profile errors are recorded without preventing the benchmark report from being generated.

Which issue(s) this PR fixes:

Fixes #1462

Bug evidence (required for bug-related PRs):

N/A

Special notes for your reviewer:

Uses the checked-in protobuf binding generated from profile.proto. The tests use captured CPU and heap profiles from the router pprof endpoint.

Validation performed:

• All 50 router benchmark tests pass
• ruff check passes
• git diff --check passes

Does this PR introduce a user-facing change?:

Integrate CPU and heap pprof analysis into router benchmark reports.

[gemini-code-assist]

Caution

The consumer version of Gemini Code Assist on GitHub has been sunset. All code review activity has officially ceased.

[volcano-sh-bot]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by:
Once this PR has been reviewed and has the lgtm label, please assign yaozengzeng for approval. For more information see the Kubernetes Code Review Process.

The full list of commands accepted by this bot can be found here.

Details

Needs approval from an approver in each of these files:

• OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

[avinxshKD]

@StLeoX PTAL when you get a chance. added the pprof parser with scheduler/plugin filtering, CPU and heap support, and fixture-based tests

cc @YaoZengzeng @LiZhenCheng9527

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

[StLeoX]

I expect the Analyzer can be integrated into build_report, instead of a separate module. You can checkout this branch first.

[avinxshKD]

@StLeoX Reworked this on top of #1285. The pprof analyzer is now integrated into ResultReporter.build_report().

[Copilot]

Pull request overview

Copilot reviewed 27 out of 29 changed files in this pull request and generated 1 comment.

Suppressed comments (1)

benchmark/router-benchmark/scripts/router_ab_test/kubernetes.py:506

• _start_port_forward() always targets "{target_type}/{self.ROUTER_DEPLOYMENT}". This happens to work today because ROUTER_DEPLOYMENT and ROUTER_SVC_NAME are both "kthena-router", but it couples service and deployment naming and will break silently if they ever diverge. Consider selecting the target resource based on target_type.

        local_endpoint = f"localhost:{local_port}"
        print(f"  Starting port-forward ({local_endpoint} → {description})")

        target = f"{target_type}/{self.ROUTER_DEPLOYMENT}"
        proc = subprocess.Popen(

[StLeoX]

lgtm, it's stacked now