[feat][evaluation] support async CustomRPC evaluators

[HearyShen]

Summary

Add asynchronous execution support for CustomRPC evaluators, primarily for long-running third-party evaluations.

• add CustomRPCEvaluator.is_async and async_invoke_http_info
• add EvaluationSPIService.AsyncInvokeEvaluator
• generalize the evaluator async coordinator to Record -> Context -> Provider
• add terminal CAS and duplicate/conflicting callback protection
• add a ResumeReady scheduling handshake
• preserve Agent evaluator async/debug metadata across callback races
• reject builtin, cross-space, and legacy-protocol async CustomRPC evaluators

Compatibility

Existing CustomRPC evaluator versions remain synchronous by default. Agent evaluator async behavior and CustomRPCServer async eval-target behavior are covered by regression tests.

Tests

• evaluation entity, converter, domain, application, and repository package tests
• critical async coordinator, callback, Redis, and metadata race tests with -race
• backend build: go build ./cmd

[codecov]

Codecov Report

❌ Patch coverage is 93.33333% with 19 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
...odules/evaluation/domain/service/evaluator_impl.go	88.54%	8 Missing and 7 partials ⚠️
...tion/infra/repo/evaluator/evaluator_record_impl.go	94.64%	2 Missing and 1 partial ⚠️
...aluation/domain/service/expt_run_item_turn_impl.go	94.44%	0 Missing and 1 partial ⚠️

@@           Coverage Diff           @@
##             main     #607   +/-   ##
=======================================
  Coverage        ?   78.04%           
=======================================
  Files           ?      689           
  Lines           ?    83986           
  Branches        ?        0           
=======================================
  Hits            ?    65551           
  Misses          ?    14542           
  Partials        ?     3893           

Flag	Coverage Δ
unittests	78.04% <93.33%> (?)

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

Files with missing lines	Coverage Δ
...ation/application/convertor/evaluator/evaluator.go	92.17% <100.00%> (ø)
...luation/application/convertor/evaluator/openapi.go	96.55% <100.00%> (ø)
...modules/evaluation/application/eval_openapi_app.go	81.63% <100.00%> (ø)
...nd/modules/evaluation/application/evaluator_app.go	90.39% <100.00%> (ø)
...kend/modules/evaluation/domain/entity/evaluator.go	100.00% <100.00%> (ø)
...tion/domain/entity/evaluator_version_custom_rpc.go	100.00% <100.00%> (ø)
...ckend/modules/evaluation/domain/entity/expt_run.go	98.64% <ø> (ø)
backend/modules/evaluation/domain/entity/param.go	87.71% <ø> (ø)
...es/evaluation/domain/service/expt_run_item_impl.go	74.60% <100.00%> (ø)
...aluation/domain/service/expt_run_item_turn_impl.go	83.70% <94.44%> (ø)
... and 2 more

---

Continue to review full report in Codecov by Harness.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update e30f3dd...eaf6510. 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.