fix(model): 解析 stop_reason + max_tokens 截断重试/非空兜底 — 不再静默空回复

- model.go: Response.StopReason; anthropic wire加stop_reason; openai finish_reason映射(两协议统一,原值透传)
- truncation.go(新,159行): effectiveMaxTokens/retryMaxTokens(×2封顶32768)/isTruncatedEmpty(空正文+无tool_use+stop_reason∈{max_tokens,length})/withEmptyBodyHint/callWithTruncationRetry/callStreamWithTruncationRetry/dropThinking/emitTextSegment
- anthropic.go Chat/ChatStream 走重试+兜底; anthropic_encode.go(新)拆编码(anthropic.go 529→357合规); anthropic_stream message_delta落StopReason
- agent/loop: 把模型层StopReason带到llm.call span的model_stop_reason属性(修掉trace硬编码stop的误导); loop.Result.StopReason语义未动
- 触发仅当截断空正文; tool_use空正文永不触发(native_toolcall_test仍过); 至多1次重试; 兜底回'（思考过长，未能生成回复）'绝不用thinking
- 新增7测试(非流式/流式重试+兜底+tool_use不触发+封顶+openai映射), build/vet/test绿+3平台交叉编译

internal/runtime/loop/step.go

@@ -17,6 +17,10 @@ type stepOutcome struct {
 	text       string
 	toolCalls  []agent.ToolCall
 	stopReason string
+	// modelStopReason is the provider's raw stop/finish reason for the model
+	// call (Anthropic max_tokens / end_turn / tool_use), recorded on the llm_call
+	// span so a truncation is visible even when the step itself ended normally.
+	modelStopReason string
 }
 
 // step runs a single model call. When an OnStream callback is configured it
@@ -45,6 +49,9 @@ func (l *Loop) step(ctx context.Context, sessionID string, messages []model.Mess
 	}
 
 	l.spanProp(llmSpan, "stop_reason", out.stopReason)
+	if out.modelStopReason != "" {
+		l.spanProp(llmSpan, "model_stop_reason", out.modelStopReason)
+	}
 	l.spanProp(llmSpan, "tool_calls", len(out.toolCalls))
 	l.finishSpan(llmSpan, trace.SpanStatusOK)
 	return out, nil
@@ -58,9 +65,10 @@ func (l *Loop) agentStep(ctx context.Context, sessionID string, messages []model
 		return nil, fmt.Errorf("loop: agent step: %w", err)
 	}
 	return &stepOutcome{
-		final:      sr.Final,
-		text:       sr.Text,
-		toolCalls:  sr.ToolCalls,
-		stopReason: sr.StopReason,
+		final:           sr.Final,
+		text:            sr.Text,
+		toolCalls:       sr.ToolCalls,
+		stopReason:      sr.StopReason,
+		modelStopReason: sr.ModelStopReason,
 	}, nil
 }

internal/runtime/loop/stream.go

@@ -53,7 +53,7 @@ func (l *Loop) streamStep(ctx context.Context, messages []model.Message) (*stepO
 	}
 
 	if calls, ok := decodeToolCalls(resp); ok {
-		return &stepOutcome{final: false, toolCalls: calls, stopReason: agent.StopReasonToolCalls}, nil
+		return &stepOutcome{final: false, toolCalls: calls, stopReason: agent.StopReasonToolCalls, modelStopReason: resp.StopReason}, nil
 	}
-	return &stepOutcome{final: true, text: resp.Content, stopReason: agent.StopReasonStop}, nil
+	return &stepOutcome{final: true, text: resp.Content, stopReason: agent.StopReasonStop, modelStopReason: resp.StopReason}, nil
 }