feat: add tool_calling_sound to BackgroundAudioPlayer

[AKomplished-bug]

Add support for separate audio feedback during function tool execution:

• Add FunctionToolsExecutingEvent to signal when tool execution starts
• Add tool_calling_sound parameter to BackgroundAudioPlayer
• Auto-switch from thinking_sound to tool_calling_sound during tool execution
• Stop tool_calling_sound when function_tools_executed fires

This allows different audio cues for LLM thinking vs tool execution, improving user feedback during voice agent interactions.

Usage

background_audio = BackgroundAudioPlayer(
    thinking_sound=BuiltinAudioClip.HOLD_MUSIC,      # During LLM processing
    tool_calling_sound=BuiltinAudioClip.KEYBOARD_TYPING,  # During tool execution
)

Changes

- events.py: Added FunctionToolsExecutingEvent class
- agent_activity.py: Emit function_tools_executing event when tool starts
- background_audio.py: Added tool_calling_sound parameter and event handlers
- __init__.py: Export FunctionToolsExecutingEvent

[CLAassistant]

All committers have signed the CLA.

[AKomplished-bug]

The failing test test_chat_ctx.py::test_summarize is unrelated to my changes. It's failing due to a missing OPENAI_API_KEY in the CI environment
(Error 401: "You didn't provide an API key"). My changes only modify events.py, agent_activity.py, background_audio.py, and init.py in the voice module. @theomonnom