[FEATURE]: Education AI Suite - System & Performance Metrics panel to display LLM/ASR configuration and performance KPIs for Smart Classroom sample app

Description Extend the System & Performance Metrics panel to include software-level metrics such as model names, inference latency, and token throughput. This helps ISVs evaluate the performance of their AI workloads on Intel platforms.

Assumptions:

• Models include Qwen2-7B, Qwenvl2-2B, and ASR models like Whisper or Conformer 1 2.
• Metrics include TTFT (Time to First Token), TPOT (Time Per Output Token), and total tokens processed.
• Metrics are collected per session and reset on new input.

Dependencies

• LLM and ASR inference engines (IPEX-LLM, OpenVINO)
• Token counters and latency timers
• GUI integration

Expected Output

• Display of:
• LLM: Qwen2-7B, Qwenvl2-2B
• ASR: [Model name]
• TTFT, TPOT, total tokens
• Time to generate summary

Acceptance Criteria

• Metrics are displayed after each inference session
• Values are accurate and updated per session
• Display is stable and readable