Preserve tool set scopes in Studio workflow documents

[AbigailDeng]

Problem

Template instantiation stores runtime-supported tool_sets in the Studio draft YAML, but Studio rejected the field as unknown during the next parse and could not save the workflow.

Solution

• Add typed ToolSets fields to Studio role and step documents.
• Accept, parse, normalize, and serialize tool_sets independently from allowed_tools.
• Preserve absent, empty, and populated lists so runtime tool-set restriction semantics remain unchanged.

Verification

• /Users/abigaildeng/.dotnet/dotnet test test/Aevatar.Studio.Tests/Aevatar.Studio.Tests.csproj --nologo --no-restore --filter "FullyQualifiedNameWorkflowCompatibilityProfileTests|FullyQualifiedNameEditorControllerSerializationTests|FullyQualifiedName~WorkflowDocumentNormalizerTests" (119 passed)
• PATH with temporary Python 3.12 python3 entrypoint bash tools/ci/test_stability_guards.sh (passed)
• git diff --check (passed)

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 83.43%. Comparing base (6781f4f) to head (ca4abd4).
⚠️ Report is 24 commits behind head on feature/integrate.

@@                  Coverage Diff                  @@
##           feature/integrate    #3506      +/-   ##
=====================================================
+ Coverage              83.39%   83.43%   +0.04%     
=====================================================
  Files                   1659     1666       +7     
  Lines                 186215   187957    +1742     
  Branches               22432    22600     +168     
=====================================================
+ Hits                  155294   156828    +1534     
- Misses                 20517    20644     +127     
- Partials               10404    10485      +81     

Flag	Coverage Δ
ci	83.43% <100.00%> (+0.04%)	⬆️

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

Files with missing lines	Coverage Δ
...udio/Compatibility/WorkflowCompatibilityProfile.cs	99.51% <100.00%> (+<0.01%)	⬆️
...c/Aevatar.Studio.Domain/Studio/Models/RoleModel.cs	100.00% <100.00%> (ø)
...c/Aevatar.Studio.Domain/Studio/Models/StepModel.cs	100.00% <100.00%> (ø)
...main/Studio/Services/WorkflowDocumentNormalizer.cs	91.41% <100.00%> (+0.10%)	⬆️

... and 16 files with indirect coverage changes

🚀 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.

[AbigailDeng]

Post-rebase verification

Rebased onto the latest feature/integrate so the current NyxID conformance source pin is included. The earlier fast-gates failure was caused by the branch being behind that baseline, not by the Studio tool_sets implementation.

• Focused Studio tests: /Users/abigaildeng/.dotnet/dotnet test test/Aevatar.Studio.Tests/Aevatar.Studio.Tests.csproj --nologo --filter "FullyQualifiedName~EditorControllerSerializationTests|FullyQualifiedName~WorkflowCompatibilityProfileTests|FullyQualifiedName~WorkflowDocumentNormalizerTests" - 119 passed, 0 failed
• Stability guards: PATH=/Users/abigaildeng/.cache/codex-runtimes/codex-primary-runtime/dependencies/python/bin:$PATH bash tools/ci/test_stability_guards.sh - passed, including 9/9 semantic tests
• NyxID conformance: PATH=/Users/abigaildeng/.cache/codex-runtimes/codex-primary-runtime/dependencies/python/bin:$PATH bash tools/ci/nyxid_conformance_guard.sh - passed: 46 top-level commands, 226 leaves, one outcome row per leaf
• Diff validation: git diff --check origin/feature/integrate...HEAD - passed