[feat][evaluation] expose target execution context for run initiators

[HearyShen]

Evaluation targets need the initiator of the specific experiment run when granting access to generated artifacts. Target-registration authors and the experiment's original creator can differ from the owner of a rerun.

Adds an authenticated, read-only execution-context endpoint. It checks workspace access before reading the persisted target record and its associated run, then returns the run initiator and execution coordinates. Missing or deleted records/runs fail closed. Shared targets retain distinct target and experiment workspace IDs.

Validation: the four affected package suites passed; focused tests and race checks cover authorization before lookup, reruns, same-run appends, shared workspaces, missing identities, and soft deletion using the real repository implementation. HTTP/application builds and vet passed. Independent code review found no blocking defects. The generated API changes were reviewed as additions with no removed definitions.

Draft pending downstream integration and end-to-end verification.

[codecov]

Codecov Report

❌ Patch coverage is 85.71429% with 8 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
...api/handler/coze/loop/apis/eval_open_apiservice.go	0.00%	2 Missing ⚠️
...ation/application/eval_target_execution_context.go	92.00%	1 Missing and 1 partial ⚠️
...luation/domain/service/target_execution_context.go	90.90%	1 Missing and 1 partial ⚠️
...s/evaluation/infra/repo/experiment/expt_run_log.go	71.42%	1 Missing and 1 partial ⚠️

@@            Coverage Diff             @@
##             main     #671      +/-   ##
==========================================
+ Coverage   78.82%   78.85%   +0.02%     
==========================================
  Files         707      709       +2     
  Lines       87823    87950     +127     
==========================================
+ Hits        69227    69349     +122     
- Misses      14607    14609       +2     
- Partials     3989     3992       +3     

Flag	Coverage Δ
unittests	78.85% <85.71%> (+0.02%)	⬆️

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

Files with missing lines	Coverage Δ
...api/handler/coze/loop/apis/eval_open_apiservice.go	4.54% <0.00%> (-0.07%)	⬇️
...ation/application/eval_target_execution_context.go	92.00% <92.00%> (ø)
...luation/domain/service/target_execution_context.go	90.90% <90.90%> (ø)
...s/evaluation/infra/repo/experiment/expt_run_log.go	83.78% <71.42%> (+7.11%)	⬆️

... and 5 files with indirect coverage changes

---

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 b699d63...a2accfb. 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.