[feat][backend] code evaluator

[tpfz]

What type of PR is this?

Check the PR title.

• This PR title match the format: [<type>][<scope>]: <description>. For example: [fix][backend] flaky fix
• The description of this PR title is user-oriented and clear enough for others to understand.
• Add documentation if the current PR requires user awareness at the usage level.
• This PR is written in English. PRs not in English will not be reviewed.

(Optional) Translate the PR title into Chinese.

(Optional) More detailed description for this PR(en: English/zh: Chinese).

en:
zh(optional):

(Optional) Which issue(s) this PR fixes:

[codecov]

Codecov Report

❌ Patch coverage is 81.67457% with 116 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
...ules/evaluation/infra/runtime/http_faas_runtime.go	68.20%	66 Missing and 3 partials ⚠️
...modules/evaluation/infra/runtime/python_runtime.go	85.22%	11 Missing and 2 partials ⚠️
...ackend/modules/evaluation/infra/runtime/factory.go	88.77%	8 Missing and 3 partials ⚠️
...les/evaluation/infra/runtime/javascript_runtime.go	88.88%	9 Missing and 1 partial ⚠️
...ackend/modules/evaluation/infra/runtime/manager.go	91.01%	6 Missing and 2 partials ⚠️
backend/modules/evaluation/pkg/conf/evaluator.go	90.00%	5 Missing ⚠️

@@                Coverage Diff                @@
##             release/v1.4.0     #220   +/-   ##
=================================================
  Coverage                  ?   64.27%           
=================================================
  Files                     ?      507           
  Lines                     ?    56715           
  Branches                  ?        0           
=================================================
  Hits                      ?    36451           
  Misses                    ?    17745           
  Partials                  ?     2519           

Flag	Coverage Δ
unittests	64.27% <81.67%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
...nd/api/handler/coze/loop/apis/evaluator_service.go	4.34% <100.00%> (ø)
backend/modules/evaluation/pkg/conf/evaluator.go	76.85% <90.00%> (ø)
...ackend/modules/evaluation/infra/runtime/manager.go	91.01% <91.01%> (ø)
...les/evaluation/infra/runtime/javascript_runtime.go	88.88% <88.88%> (ø)
...ackend/modules/evaluation/infra/runtime/factory.go	88.77% <88.77%> (ø)
...modules/evaluation/infra/runtime/python_runtime.go	85.22% <85.22%> (ø)
...ules/evaluation/infra/runtime/http_faas_runtime.go	68.20% <68.20%> (ø)

---

Continue to review full report in Codecov by Sentry.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update cab8bde...6cab99d. Read the comment docs.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[CozeLoop]

Summary:

• Must-Fix: JS return_val is not captured by the current JS FaaS wrapper because it logs a raw string. Please emit a single JSON line (e.g., console.log(JSON.stringify({ret_val}))) so the server can parse it, or add an explicit marker protocol.
• Should-Fix: Provide fail-open defaults for COZE_LOOP_JS_FAAS_URL and COZE_LOOP_PYTHON_FAAS_URL to improve robustness across dev/test environments.
• Suggestion: Use normalizeLanguage(language) in the HTTP adapter, and include TypeScript aliases (ts/typescript) in config key normalization.

These changes will restore functional correctness for JS evaluators, reduce configuration friction, and make the runtime adapter more resilient.

[CozeLoop]

Aime CR summary: Backend runtime integration (factory/manager + HTTP FaaS adapter) is correct and concurrency-safe. ValidateEvaluator handler correctly restores service invocation. Config decoding and language normalization look good. No Must-Fix or Should-Fix issues found on backend Go changes. A few non-blocking suggestions: centralize default FaaS BaseURL config, remove unused basicSyntaxValidation or wire it into ValidateCode, and consider unifying logs to English for consistency. Full report is available in Feishu: https://bytedance.larkoffice.com/docx/RbfJdgpZsoqiBhxJ6epcqYWXnQf