feat: add auto-instrumentation to langchain agent template

[avi-alpert]

Description

Adds auto-instrumentation to the LangChain/LangGraph agent templates so that traces emitted by LangChain agents are compatible with AgentCore's observability and evaluation features out of the box.

Changes:

• Import and call LangchainInstrumentor().instrument() at module level in both A2A and HTTP LangChain template main.py files
• Add opentelemetry-instrumentation-langchain to the template pyproject.toml dependencies for both A2A and HTTP variants
• Update asset snapshots

Why: Without the LangChain OpenTelemetry instrumentor, traces from LangChain agents are emitted under a generic scope that the Evaluate API does not recognize. Running agentcore run eval against a LangChain agent would fail with:

Error: Provided input has no spans with supported scope. Currently supported scopes are ['strands.telemetry.tracer', 'opentelemetry.instrumentation.langchain', 'openinference.instrumentation.langchain']. Please verify that you are using the supported agentic-framework and instrumentation-library

With this change, LangChain agent traces are emitted under the opentelemetry.instrumentation.langchain scope, and agentcore run eval succeeds.

Reference: Configuring custom observability in AgentCore

Type of Change

• New feature

Testing

Verified by running agentcore run eval against a LangChain agent — evaluation completes successfully instead of returning the unsupported-scope error.

• I ran npm run test:unit and npm run test:integ
• I ran npm run typecheck
• I ran npm run lint
• If I modified src/assets/, I ran npm run test:update-snapshots and committed the updated snapshots

Checklist

• I have read the CONTRIBUTING document
• I have added any necessary tests that prove my fix is effective or my feature works
• I have updated the documentation accordingly
• I have added an appropriate example to the documentation to outline the feature, or no new docs are needed
• My changes generate no new warnings
• Any dependent changes have been merged and published

---

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the
terms of your choice.

[jariy17]

Every other third-party dep in this file uses a >= X.Y.Z minimum pin (langgraph >= 0.2.0, bedrock-agentcore >= 1.0.3, etc.). This package is from the Traceloop/OpenLLMetry project (not AWS-controlled) and has a very active release cadence in the 0.x series. An unpinned dep could pull a breaking version for users. Same applies to the HTTP variant.

Suggested change

	"opentelemetry-instrumentation-langchain",
	"opentelemetry-instrumentation-langchain >= 0.38.0",