feat:add_request_level_token_callback

[fengju0213]

Related Issue

Closes #https://github.com/eigent-ai/eigent_server/issues/46

Description

Summary

This PR adapts Eigent to CAMEL request-level usage reporting for non-stream
mode by introducing a dedicated SSE event request_usage, enabling real-time
token accumulation during a step loop.

Background

Previously, Eigent consumed token usage mainly via deactivate_agent (step-
level). This is too coarse for long multi-request loops inside a single step
and can delay downstream billing/accounting actions.

What Changed

• Added new backend action/event:
  • Action.request_usage
  • typed payload ActionRequestUsageData
• Wired CAMEL callback in ListenChatAgent:
  • auto-detects whether CAMEL ChatAgent supports on_request_usage
  • when available, emits request_usage per model request with:
    • tokens (request total tokens)
    • request_index
    • response_id
    • step_total_tokens
    • agent/task identity fields
• Added SSE forwarding:
  • chat_service now emits request_usage
• Frontend adaptation:
  • added AgentStep.REQUEST_USAGE
  • chatStore consumes request_usage and calls addTokens(...)
    immediately
  • updated agent message type fields for new payload keys
• Added dedup logic for non-stream mode:
  • when request-level reporting is enabled, non-stream
    deactivate_agent.tokens is set to 0
  • prevents double counting (request_usage + deactivate_agent)
• Added unit test:
  • verifies request_usage increments tokens in real time
  • verifies subsequent deactivate_agent(tokens=0) does not over-count

Compatibility

• If CAMEL version does not expose on_request_usage, Eigent falls back to
  previous step-level behavior automatically.
• Streaming behavior remains unchanged.

Validation

• python3 -m py_compile backend/app/agent/listen_chat_agent.py backend/app/ service/task.py backend/app/service/chat_service.py
• npm run test -- test/unit/store/chatStore.test.ts
• Result: passed

Scope

• Non-stream request-level token reporting only.
• No token-by-token streaming settlement in this PR.

Testing Evidence (REQUIRED)

• I have included human-verified testing evidence in this PR.
• This PR includes frontend/UI changes, and I attached screenshot(s) or screen recording(s).
• No frontend/UI changes in this PR.

What is the purpose of this pull request?

• Bug fix
• New Feature
• Documentation update
• Other

Contribution Guidelines Acknowledgement

• I have read and agree to the Eigent Contribution Guideline

[fengju0213]

Screen.Recording.2026-02-24.at.14.47.55.mp4

The video shows that the token is updated with each tool call (request).

[fengju0213]

pending camel-ai/camel#3872

[Wendong-Fan]

seems in streaming mode _stream_chunks / _astream_chunks call _send_agent_deactivate(accumulated_content, total_tokens) in their finally block with the full step token count, while request_usage events have already reported those same tokens, the frontend adds both, resulting in inflated token counts?

added pr #1439