add name attribute to FunctionBaseConfig for workflow naming in span exporter

[bbednarski9]

Description

This PR adds the name attribute to FunctionBaseConfig, enabling workflows to be optionally named (non-breaking, backwards compatible). This allows developers to optionally name workflows and have those names carried through the span exporter and represented in OTEL data. This PR follows #1320 which enabled span parent-child lineage for nested tool calls. /examples/observability/simple_calculator_observability/config-phoenix-nested.yml has been updated - showing this working in practice.
Closes
#1335

By Submitting this PR I confirm:

• I am familiar with the Contributing Guidelines.
• We require that all contributors "sign-off" on their commits. This certifies that the contribution is your original work, or you have rights to submit it under the same license, or a compatible license.
  • Any contribution which contains commits that are not Signed-Off will not be accepted.
• When the PR is ready for review, new or existing tests cover these changes.
• When the PR is ready for review, the documentation is up to date with these changes.

Summary by CodeRabbit

• New Features
  • Optional display names for functions and workflows used in tracing and observability, improving monitoring visibility; sensible identifiers are derived automatically when omitted.
• Refactor
  • Enhanced workflow name-resolution for run and streaming events so observability consistently shows meaningful names (explicit name → instance → type → fallback).
• Tests
  • Added unit tests validating the new name-resolution behavior for observability events.
• Chores
  • Disabled a local pre-commit hook.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

[coderabbitai]

Walkthrough

Adds an optional name field to function/workflow configs and updates builder and runner logic to avoid exposing the workflow placeholder by preferring config.name, then instance_name, and finally config.type; includes tests, an example config update, and disables a local pre-commit hook.

Changes

Cohort / File(s)	Summary
Observability config examples/observability/simple_calculator_observability/configs/config-phoenix-nested.yml	Added workflow.name set to power_of_two_agent.
Function schema src/nat/data_models/function.py	Added optional `name: str
Builder instance naming src/nat/builder/function.py	Imported WORKFLOW_COMPONENT_NAME; changed instance_name initialization to avoid exposing the workflow placeholder and to prefer config.name or config.type when appropriate.
Runtime name resolution src/nat/runtime/runner.py	Imported WORKFLOW_COMPONENT_NAME; reworked workflow_name resolution for result and streaming paths to prefer config.name, then instance_name, treat placeholder instance_name by using config.type, and allow None if unresolved.
Tests tests/nat/runtime/test_runner_trace_ids.py	Added test_runner_workflow_name_resolution to verify resolution across config.name, instance_name, and config.type fallbacks; added test scaffolding and patches; removed an asyncio marker from an existing test.
Pre-commit config .pre-commit-config.yaml	Commented out the local clear-notebook-output-cells hook, disabling it.
Manifests / Metadata manifest_file, pyproject.toml, requirements.txt	Minor manifest references noted in summary; no functional manifest edits in the diff.

Sequence Diagram(s)

sequenceDiagram
    participant Config as Config (workflow/function)
    participant Builder as Builder
    participant Runtime as Runner
    participant Tracing as Tracing/Exporter

    Config->>Builder: provide function config (may include `name`)
    Builder->>Builder: determine `instance_name`
    note right of Builder: prefer provided instance_name\nif == WORKFLOW_COMPONENT_NAME use config.name → config.type
    Builder->>Runtime: supply Function with `instance_name` and config
    Runtime->>Runtime: resolve `workflow_name`
    note right of Runtime: prefer config.name → instance_name\nif instance_name == WORKFLOW_COMPONENT_NAME then use config.type\nelse allow None
    Runtime->>Tracing: emit WORKFLOW_START with resolved `workflow_name`
    Runtime->>Tracing: emit WORKFLOW_END with resolved `workflow_name`

Loading

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 10.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately describes the main change: adding a name attribute to FunctionBaseConfig for workflow naming. However, it exceeds the recommended 72-character limit at 77 characters.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

• 📝 Generate docstrings

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In `@src/nat/data_models/function.py`:
- Around line 27-38: Update the docstring in the Function configuration class to
wrap code entities in backticks: change occurrences of name and middleware to
`name` and `middleware`, and wrap the literal `middleware` section reference and
`YAML` if mentioned, so all code/config entities are backticked; locate the
docstring above the Field declaration for the name attribute (and the comment
block describing middleware) and apply the backticks consistently for `name`,
`middleware`, and any inline YAML section references.

🧹 Nitpick comments (1)

src/nat/runtime/runner.py (1)

168-180: Deduplicate workflow name resolution logic.

The same fallback chain appears in both result and result_stream. Centralizing it reduces drift and keeps behavior consistent.

♻️ Suggested refactor

 class Runner:
+    def _resolve_workflow_name(self) -> str:
+        """Resolve workflow name with backward-compatible fallback chain."""
+        config = getattr(self._entry_fn, "config", None)
+        workflow_name = getattr(config, "name", None)
+        if not workflow_name:
+            workflow_name = getattr(self._entry_fn, "instance_name", None)
+            if workflow_name == WORKFLOW_COMPONENT_NAME:
+                workflow_name = getattr(config, "type", None) or "workflow"
+            elif not workflow_name:
+                workflow_name = "workflow"
+        return workflow_name
@@
-            config = getattr(self._entry_fn, 'config', None)
-            workflow_name = getattr(config, 'name', None)
-            if not workflow_name:
-                workflow_name = getattr(self._entry_fn, 'instance_name', None)
-                if workflow_name == WORKFLOW_COMPONENT_NAME:
-                    workflow_name = getattr(config, 'type', None) or "workflow"
-                elif not workflow_name:
-                    workflow_name = "workflow"
+            workflow_name = self._resolve_workflow_name()
@@
-            config = getattr(self._entry_fn, 'config', None)
-            workflow_name = getattr(config, 'name', None)
-            if not workflow_name:
-                workflow_name = getattr(self._entry_fn, 'instance_name', None)
-                if workflow_name == WORKFLOW_COMPONENT_NAME:
-                    workflow_name = getattr(config, 'type', None) or "workflow"
-                elif not workflow_name:
-                    workflow_name = "workflow"
+            workflow_name = self._resolve_workflow_name()

Also applies to: 260-272

[willkill07]

Overall adding name makes sense. Would "display_name" be less likely to conflict with a user's existing configuration?

[bbednarski9]

/ok to test 898d116

[bbednarski9]

/ok to test 898d116

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In `@tests/nat/runtime/test_runner_trace_ids.py`:
- Line 122: Remove the unnecessary `@pytest.mark.asyncio` decorator from the async
test in tests/nat/runtime/test_runner_trace_ids.py (the test currently annotated
with `@pytest.mark.asyncio`); leave the async def test_... function as-is since
pytest-asyncio auto-detects async tests, and run the test suite to confirm
behavior remains unchanged.

[bbednarski9]

/ok to test 0695894

[bbednarski9]

/ok to test 0b45e99

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In `@tests/nat/runtime/test_runner_trace_ids.py`:
- Around line 141-145: The unused parameter to_type in the methods convert and
ainvoke should be explicitly marked as intentionally unused to silence Ruff
ARG002; inside each method (convert and ainvoke) add a simple no-op use of the
parameter (e.g., assign it to _ or an unused variable) with a brief comment like
"# silence ruff ARG002" so the signature stays the same but the linter sees the
parameter as used.

[bbednarski9]

/ok to test 0b45e99

[bbednarski9]

/merge