修复: token 汇总正确性(会话级双计/is_estimated跨端/流式usage传递) + is_estimated SELECT 缺失修复

crates/df-ai/src/anthropic_helpers.rs

@@ -165,7 +165,7 @@ pub(crate) fn apply_anthropic_event(data: &str, usage_accum: &mut Option<TokenUs
                     reasoning_tokens: 0,
                 });
             }
-            StreamChunk { delta: String::new(), finished: false, tool_calls: None, usage: None, error: None, reasoning_content: None }
+            StreamChunk { delta: String::new(), finished: false, tool_calls: None, usage: usage_accum.clone(), error: None, reasoning_content: None }
         }
         // 消息增量：output_tokens 是累计值（非增量），直接覆盖 completion + 重算 total
         "message_delta" => {
@@ -173,9 +173,11 @@ pub(crate) fn apply_anthropic_event(data: &str, usage_accum: &mut Option<TokenUs
                 let acc = usage_accum
                     .get_or_insert(TokenUsage::default());
                 acc.completion_tokens = out as u32;
-                acc.total_tokens = acc.prompt_tokens + acc.completion_tokens;
+                acc.total_tokens = acc.prompt_tokens.saturating_add(acc.completion_tokens);
             }
-            StreamChunk { delta: String::new(), finished: false, tool_calls: None, usage: None, error: None, reasoning_content: None }
+            // 加固:usage 挂到本帧(而非仅 message_stop 带出),中途断连/端点不发 message_stop
+            // 时仍能拿到真实 usage,对称 openai_helpers 的修复。
+            StreamChunk { delta: String::new(), finished: false, tool_calls: None, usage: usage_accum.clone(), error: None, reasoning_content: None }
         }
         // 文本增量
         "content_block_delta" => {

crates/df-ai/src/openai_helpers.rs

@@ -256,17 +256,19 @@ pub(crate) fn apply_openai_sse(data: &str, usage_accum: &mut Option<TokenUsage>)
                     delta: delta_text,
                     finished,
                     tool_calls,
-                    usage: None,
+                    // 加固:usage 与 finish_reason 同帧的端点(部分兼容实现),此处已累积则挂上,
+                    // 使下游 stream_recv 不必等到 [DONE] 帧即可拿到真实 usage。
+                    usage: usage_accum.clone(),
                     error: None,
                     reasoning_content: choice.delta.reasoning_content,
                 }
             } else {
-                // choices 为空 = usage-only chunk，不输出文本（usage 已累积）
+                // choices 为空 = usage-only chunk，不输出文本（usage 已累积），usage 一并带出
                 StreamChunk {
                     delta: String::new(),
                     finished: false,
                     tool_calls: None,
-                    usage: None,
+                    usage: usage_accum.clone(),
                     error: None,
                     reasoning_content: None,
                 }