feat: add LangChain LangGraph adapter MVP

[mertsatilmaz]

Fixes #51

Describe the changes you have made in this PR

Adds an opt-in langchain adapter so the harness can drive LangChain Runnables
and LangGraph apps as targets, alongside the existing in-process adapter.

• New src/agent_harness/langchain_adapter.py:
  • build_langchain_input serializes the scenario payload into the
    {"messages": [...]} state shape both Runnables and compiled LangGraphs accept.
  • load_langchain_target resolves module:object import paths and surfaces
    a clear AdapterError when the optional deps are missing.
  • Streams the run and converts LangChain message/tool events into the
    harness's vendor-neutral Trace via TraceRecorder.
• runner.py and cli.py wired to dispatch to the new adapter when
  --adapter langchain is passed; behavior for the default adapter is unchanged.
• pyproject.toml: adds a [langchain] extra (langchain + langgraph). Core
  stays vendor-neutral — nothing imported unless the adapter is selected.
• Example target at examples/targets/langchain_runnable_agent.py.
• Docs: docs/adapters.md updated with install + usage; README cross-links it.

---

Code Understanding and AI Usage

Did you use AI assistance (ChatGPT, Claude, Copilot, etc.) to write any part of this code or documentation?

• No, I wrote all the code myself
• Yes, I used AI assistance

If you used AI assistance, briefly describe:

• Tool(s) used:
• Parts of the contribution that were AI-assisted:
• How you reviewed the output:
• Tests or checks I ran:

---

Checklist before requesting a review

• I have added proper PR title and linked to the issue
• I have performed a self-review of my code
• I can explain the purpose of every documentation, function, class, and logic block I added
• I understand why my changes work and have tested them thoroughly
• I have considered potential edge cases and how my code handles them
• If it is a core feature, I have added thorough tests
• My code follows the project's style guidelines and conventions

Test plan

• pytest tests/test_langchain_adapter.py — covers input shaping, import-path
  parsing, missing-deps error, and trace conversion.
• pytest tests/test_cli.py — covers --adapter langchain dispatch and error
  paths.
• Manual: agent-harness run … --adapter langchain --target examples.targets.langchain_runnable_agent:agent