feat: fix the tool mock name and add the Evaluation Output Span

[AAgnihotry]

Summary

• Add is_tool_simulated(tool_name: str) -> bool function that checks whether a tool will be simulated based on the current evaluation context
• Add _normalize_tool_name helper to handle underscore/space normalization in tool names (e.g., my_tool matches config with my tool)
• Export is_tool_simulated from uipath.eval.mocks public API
• Add "Evaluation output" span as a child of each evaluator span in evaluation tracing containing:
  • value: the evaluation score
  • justification: extracted from evaluation details
  • evaluatorId: the ID of the evaluator
• Bump version to 2.4.8

Test plan

• Run existing tests to ensure no regressions
• Add unit tests for _normalize_tool_name function
• Add unit tests for is_tool_simulated with various scenarios:
  • No evaluation context
  • Null mocking strategy
  • LLM strategy with simulated/non-simulated tools
  • Mockito strategy with simulated/non-simulated tools
  • Underscore/space normalization handling
• Add tests for new "Evaluation output" span:
  • Span name and type
  • Required attributes (value, evaluatorId, justification)
  • Span hierarchy (child of evaluator)
  • Justification extraction from different detail formats

🤖 Generated with Claude Code

Development Package

• Use uipath pack --nolock to get the latest dev build from this PR (requires version range).
• Add this package as a dependency in your pyproject.toml:

[project]
dependencies = [
  # Exact version:
  "uipath==2.4.8.dev1010783672",

  # Any version from PR
  "uipath>=2.4.8.dev1010780000,<2.4.8.dev1010790000"
]

[[tool.uv.index]]
name = "testpypi"
url = "https://test.pypi.org/simple/"
publish-url = "https://test.pypi.org/legacy/"
explicit = true

[tool.uv.sources]
uipath = { index = "testpypi" }

[tool.uv]
override-dependencies = [
    "uipath>=2.4.8.dev1010780000,<2.4.8.dev1010790000",
]

related PR

https://github.com/UiPath/uipath-agents-python/pull/79