Phase 1 & 2: Baseline experiments + AGoT implementation + reasoning graph analysis

## Added

### Core framework
- src/llm_client.py — Ollama API wrapper for local Docker inference
- src/samples.py — Embedded benchmark datasets (GSM8K/MATH/AIME + Chinese)
- src/run_all.py — Unified experiment runner

### Phase 1: Baselines
- src/baseline/benchmark.py — IO, CoT, CoT-SC, ToT evaluation with Chinese support

### Phase 2: AGoT
- src/agot/agot.py — AGoT core algorithm (6 agent types, recursive decomposition)
- src/agot/graph_utils.py — DAG graph data structure with cycle detection
- src/agot/prompts.py — 6 agent prompt templates (EN + ZH)
- src/agot/run.py — AGoT experiment runner

### Graph Analysis
- src/graph_analysis/reasoning_graph.py — Graph property computation (cyclicity, diameter, small-world)
- src/graph_analysis/visualize.py — Visualization charts

### Documentation
- docs/graph_reasoning_principles.md — Comprehensive principles document
- docs/next_steps.md — Roadmap for reliable reasoning tool

### Experiment Results
- GSM8K/MATH baselines on qwen2.5:32b and qwq:latest
- AGoT validation (100% on small sample)
- Chinese dataset benchmarks (C-GSM8K, CMATH)
- Graph property analysis confirming Topology of Reasoning findings

src/llm_client.py

@@ -0,0 +1,113 @@
+"""
+Ollama LLM Client — wraps the Ollama REST API running in Docker.
+
+Provides a unified interface for all experiment scripts to call local models.
+"""
+
+import json
+import time
+from typing import Optional
+import ollama
+
+# ── Default Configuration ─────────────────────────────────────────────
+DEFAULT_BASE_URL = "http://localhost:11434"
+DEFAULT_MODEL = "qwen2.5:32b"
+
+
+class LLMClient:
+    """A lightweight wrapper around the Ollama API."""
+
+    def __init__(
+        self,
+        model: str = DEFAULT_MODEL,
+        base_url: str = DEFAULT_BASE_URL,
+        temperature: float = 0.7,
+        top_p: float = 0.9,
+        max_tokens: int = 2048,
+        num_ctx: int = 8192,
+        seed: Optional[int] = None,
+    ):
+        self.model = model
+        self.client = ollama.Client(host=base_url)
+        self.options = dict(
+            temperature=temperature,
+            top_p=top_p,
+            num_predict=max_tokens,
+            num_ctx=num_ctx,
+        )
+        if seed is not None:
+            self.options["seed"] = seed
+
+    # ── Public API ────────────────────────────────────────────────────
+
+    def generate(self, prompt: str, system: Optional[str] = None) -> str:
+        """Simple text generation (no chat history)."""
+        messages = []
+        if system:
+            messages.append({"role": "system", "content": system})
+        messages.append({"role": "user", "content": prompt})
+        return self._chat(messages)
+
+    def chat(self, messages: list) -> str:
+        """Multi-turn chat. messages = [{"role": "user"/"assistant", "content": ...}]."""
+        return self._chat(messages)
+
+    def generate_with_metadata(self, prompt: str, system: Optional[str] = None) -> dict:
+        """Like generate() but returns full response metadata."""
+        messages = []
+        if system:
+            messages.append({"role": "system", "content": system})
+        messages.append({"role": "user", "content": prompt})
+        return self._chat_full(messages)
+
+    def count_tokens(self, text: str) -> int:
+        """Estimate token count (rough: ~4 chars per token for Chinese+English)."""
+        return len(text) // 3 + 1
+
+    @property
+    def model_name(self) -> str:
+        return self.model
+
+    # ── Internal ──────────────────────────────────────────────────────
+
+    def _chat(self, messages: list) -> str:
+        resp = self._chat_full(messages)
+        return resp["message"]["content"]
+
+    def _chat_full(self, messages: list) -> dict:
+        resp = self.client.chat(
+            model=self.model,
+            messages=messages,
+            options=self.options,
+        )
+        return resp
+
+    def __repr__(self) -> str:
+        return f"LLMClient(model={self.model})"
+
+
+# ── Convenience Factory ───────────────────────────────────────────────
+
+def make_client(
+    model: str = DEFAULT_MODEL,
+    temperature: float = 0.0,
+    max_tokens: int = 2048,
+    seed: int = 42,
+) -> LLMClient:
+    """Create a deterministic client (temperature=0) suitable for evaluation."""
+    return LLMClient(
+        model=model,
+        temperature=temperature,
+        max_tokens=max_tokens,
+        seed=seed,
+    )
+
+
+# ── Quick Test ────────────────────────────────────────────────────────
+
+if __name__ == "__main__":
+    # Test the client
+    client = make_client()
+    print(f"Model: {client.model_name}")
+    resp = client.generate("Say hello in exactly one word.")
+    print(f"Response: {resp}")