feat(examples): use Chinese ecommerce agent names

[sipercai]

Description

This is a follow-up to #260. It changes the two ecommerce customer-service agent display names from English to Chinese so that traces for Chinese users show:

• invoke_agent 售前服务
• invoke_agent 售后服务

The LangGraph node names, Python identifiers, route values, and tool names remain in English to keep the implementation stable and easy to maintain.

The English and Chinese READMEs now document the expected trace names.

Fixes # (N/A)

Type of change

• New feature (non-breaking change which adds functionality)
• This change requires a documentation update

How Has This Been Tested?

• cd examples/ecommerce-customer-service && uv run --isolated --no-project --with-requirements requirements.txt --with-requirements requirements-dev.txt python -m pytest tests -q (10 passed)
• tox -e precommit
• Local telemetry smoke verified both Chinese invoke_agent span names and gen_ai.agent.name values
• Provider-backed deployment smoke covered presales, aftersales, and clarification requests
• Cloud trace readback verified the specialist, ReAct model/tool, business-tool, and reviewer span chain
• Independent diff review completed with zero actionable findings

Validation Evidence

Area	Result	Evidence / reason
Scope	Pass	Example and its tests/docs only; no instrumentation package changed
Focused tests	Pass	10 tests passed
Pre-commit	Pass	All repository hooks passed from a fresh cache
Test-first regression	N/A	This is an example display-name configuration change, not an instrumentation bug fix
Provider VCR	N/A	No provider client or instrumentation behavior changed
Non-streaming workflow	Pass	Presales, aftersales, and clarification paths returned successfully
Streaming workflow	N/A	This example intentionally uses the non-streaming invoke workflow
Concurrent request isolation	Pass	Existing concurrency test passed
Agent/tool/ReAct flow	Pass	Both specialist agents invoked their expected tools before review
Error/fail-open path	Pass	Existing error-path tests passed
Trace tree	Pass	Chinese specialist span names and agent attributes were observed while English graph nodes remained unchanged
Content capture	N/A	No content-capture behavior changed
Weaver semantic check	N/A	No semantic-convention registry or instrumentation mapping changed; the existing exported contract was checked by cloud trace readback
GitHub CI	Pending	CI starts after the PR is published

Does This PR Require a Core Repo Change?

• Yes. - Link to PR:
• No.

Checklist:

• Followed the style guidelines of this project
• Changelogs have been updated
• Unit tests have been added
• Documentation has been updated

The PR should carry the Skip Changelog label because it only adjusts an example's user-facing agent names and documentation.

[ralf0131]

LGTM. Clean, focused change — introduces named constants for the Chinese agent display names (售前服务 / 售后服务) and keeps the English identifiers stable. Documentation updates in both READMEs accurately describe the expected trace span names. Tests properly reference the constants instead of hardcoded strings.

---

Automated review by github-manager-bot

[ralf0131]

LGTM. Trivial change, well-tested (10 tests pass), documentation updated in both EN and ZH. Good use of named constants for maintainability.

---

Automated review by github-manager-bot