Store headstage sensor data as separate TimeSeries objects with appropriate units and scaling

[Copilot]

This PR implements separate storage of headstage sensor data (accelerometer, gyroscope, magnetometer) as individual TimeSeries objects in the acquisition module with proper physical units and scaling factors, addressing the issue where all analog data was previously combined into a single TimeSeries with generic units.

Key Changes

Enhanced Analog Data Processing:

• Automatically detects and categorizes headstage sensor channels by type based on naming patterns:
  • Headstage_AccelX/Y/Z → Accelerometer data scaled by 0.000061 with unit g
  • Headstage_GyroX/Y/Z → Gyroscope data scaled by 0.061 with unit d/s
  • Headstage_MagX/Y/Z → Magnetometer data with unit unspecified
  • ECU_Ain* → Analog input channels with configurable units

Metadata Schema Extensions:

• Added optional sensor_units field for custom unit specifications per sensor type
• Enhanced behavioral_events items to support individual unit and comments fields
• Maintains full backwards compatibility with existing metadata files

Output Structure:

• New default behavior: Creates separate TimeSeries objects in nwbfile.acquisition with descriptive names and proper units
• Legacy compatibility: Optional separate_sensor_data=False parameter preserves original combined TimeSeries in processing module

Example Usage:

# Metadata configuration
metadata = {
    "sensor_units": {
        "accelerometer": "g", 
        "gyroscope": "d/s"
    }
}

# Results in separate acquisition TimeSeries:
add_analog_data(nwbfile, rec_files, metadata=metadata)
# - "accelerometer": scaled accelerometer data in g units
# - "gyroscope": scaled gyroscope data in d/s units  
# - "magnetometer": raw magnetometer data
# - "ecu_analog_input": ECU analog channels

Benefits:

• Proper physical units applied automatically (accelerometer in g, gyroscope in degrees/second)
• Clear data organization with separate, descriptive TimeSeries objects
• Accurate scaling from raw integer values to meaningful physical measurements
• Improved NWB compliance by storing acquisition data in the acquisition module
• Backwards compatibility for existing workflows

Testing:

• Added comprehensive unit tests for sensor categorization and TimeSeries creation
• Integration tests validating full conversion pipeline with both new and legacy modes
• Tests confirm proper scaling factors and unit assignments

Fixes #19.

---

💬 Share your feedback on Copilot coding agent for the chance to win a $200 gift card! Click here to start the survey.

[Copilot]

@edeno 👋 This repository doesn't have Copilot instructions. With Copilot instructions, I can understand the repository better, work faster and produce higher quality PRs.

I can generate a .github/copilot-instructions.md file for you automatically. Click here to open a pre-filled issue and assign it to me. I'll write the instructions, and then tag you for review.

[edeno]

What happens if this isn't present in the yaml?

[edeno]

should be optional type