feat: Otel instrumentation version 3

[bitnahian]

Fixes #2964

Key Changes:

• New InstrumentationNames class - Centralizes span naming and attribute configuration for different instrumentation versions (v2 vs v3+)
• This was necessary because it didn't feel quite right to include this functionality in InstrumentationSettings. These could potentially be staticmethods in InstrumentationSettings. Feel free to suggest changing to this if it feels more appropriate.
• Versioned span naming - v3+ uses semantic naming like invoke_agent {name} and execute_tool {name} vs v2's generic names
• Updated attribute schemas - v3+ uses standardized gen_ai.* attributes (e.g., gen_ai.agent.name, gen_ai.tool.call.arguments)
• Output tool instrumentation - Added proper span configuration for output function execution

[Copilot]

Pull Request Overview

This PR adds support for OpenTelemetry instrumentation version 3, implementing a new configuration system for span naming and attribute schemas. The main purpose is to provide semantic span names (e.g., invoke_agent {name}, execute_tool {name}) and standardized gen_ai.* attributes for better observability.

• Added InstrumentationConfig class to centralize span naming and attribute configuration across versions
• Updated span naming patterns for v3+ with semantic naming conventions
• Migrated attribute schemas to use standardized gen_ai.* format for v3+

Reviewed Changes

Copilot reviewed 9 out of 9 changed files in this pull request and generated 4 comments.

Show a summary per file

File	Description
pydantic_ai_slim/pydantic_ai/_instrumentation.py	New file implementing InstrumentationConfig class for version-specific span naming and attributes
pydantic_ai_slim/pydantic_ai/models/instrumented.py	Updated version type annotation to include version 3 and changed default to use constant
pydantic_ai_slim/pydantic_ai/agent/init.py	Updated agent run span creation to use versioned configuration
pydantic_ai_slim/pydantic_ai/_tool_manager.py	Modified tool execution spans to use versioned naming and attributes
pydantic_ai_slim/pydantic_ai/_output.py	Updated output function spans to use versioned configuration
pydantic_ai_slim/pydantic_ai/_run_context.py	Added instrumentation version field to RunContext
pydantic_ai_slim/pydantic_ai/_agent_graph.py	Updated to pass instrumentation version to run context
tests/test_logfire.py	Comprehensive test updates with version 3 test cases and expected span names/attributes
tests/models/test_fallback.py	Added gen_ai.agent.name attribute to test expectations

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[Copilot]

Pull Request Overview

Copilot reviewed 9 out of 9 changed files in this pull request and generated 4 comments.

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[alexmojaki]

Thanks!

[bitnahian]

Thanks @alexmojaki legend.