Add tool wrapper support for workflow deployments

[devin-ai-integration]

Extends the @tool(inputs={...}, examples=[...]) decorator pattern (from PR #3398) to support DeploymentDefinition objects in ToolCallingNode. This allows workflow deployments to specify which parameters come from parent workflow inputs vs the LLM, matching the existing behavior for inline workflows and function tools.

---

Review & Testing Checklist for Human

• Verify serialization output matches platform expectations: The serialization test mocks the deployment release info - manually test that the definition field structure is what the Vellum platform expects
• Test with a real workflow deployment: The runtime test mocks the workflow execution - verify that inputs are actually merged and passed correctly to the deployment API call
• Check edge case when display_context.client is None: The code skips adding the definition in this case (line 499 in expressions.py) - verify this doesn't cause downstream issues
• Verify the function name transformation: The code removes hyphens from deployment names (e.g., weather-workflow-deployment → weatherworkflowdeployment) - confirm this is the expected behavior

Notes

• Requested by: @vincent0426
• Session: https://app.devin.ai/sessions/10589f2cd4e64b669f5346d51afb91fd

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR that start with 'DevinAI' or '@devin'.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment and CI monitoring