agent/context: make Thinking blocks prompt-visible

Thinking blocks used to render as empty strings and be excluded from
is_prompt_visible, so the model never saw its own prior CoT across
turns. For Qwen 3.6 native thinking mode, CoT is meant to stay in the
conversation — the model benefits from seeing what it reasoned about
last turn.

Render Thinking as <think>\n{text}\n</think>\n so past reasoning is
visible in subsequent prompts. Add in_think param to ResponseParser::new
so the parser starts inside a <think> block when the prompt was
prefilled with "<think>\n" (native thinking mode).

Co-Authored-By: Proof of Concept <poc@bcachefs.org>

src/agent/context.rs

@@ -218,7 +218,11 @@ impl NodeBody {
     fn render_into(&self, out: &mut String) {
         match self {
             Self::Content(text)   => out.push_str(text),
-            Self::Thinking(_)     => {},
+            Self::Thinking(text)  => {
+                out.push_str("<think>\n");
+                out.push_str(text);
+                out.push_str("\n</think>\n");
+            }
             Self::Log(_)          => {},
             Self::ToolCall { name, arguments } => {
                 out.push_str("<tool_call>\n");
@@ -258,7 +262,7 @@ impl NodeBody {
     }
 
     fn is_prompt_visible(&self) -> bool {
-        !matches!(self, Self::Thinking(_) | Self::Log(_))
+        !matches!(self, Self::Log(_))
     }
 
     /// Hand-assemble token IDs for body types where running the tokenizer
@@ -648,13 +652,17 @@ fn drain_safe(buf: &mut String, tag_len: usize) -> String {
 }
 
 impl ResponseParser {
-    pub fn new(branch_idx: usize) -> Self {
+    /// @in_think: whether the model's output begins inside a <think> block.
+    /// Set when the prompt was prefilled with "<think>\n" (native thinking
+    /// mode) so the parser captures reasoning tokens as Thinking until the
+    /// model emits </think>.
+    pub fn new(branch_idx: usize, in_think: bool) -> Self {
         Self {
             branch_idx,
             call_counter: 0,
             buf: String::new(),
             content_parts: Vec::new(),
-            in_think: false,
+            in_think,
             think_buf: String::new(),
             in_tool_call: false,
             tool_call_buf: String::new(),
@@ -1369,7 +1377,7 @@ mod tests {
     fn parse_into_ctx(chunks: &[&str]) -> (ContextState, Vec<PendingToolCall>) {
         let mut ctx = ContextState::new();
         ctx.push_no_log(Section::Conversation, AstNode::branch(Role::Assistant, vec![]));
-        let mut p = ResponseParser::new(0);
+        let mut p = ResponseParser::new(0, false);
         let mut calls = Vec::new();
         for chunk in chunks {
             // Feed each chunk as a single token (id=0 for tests)
@@ -1433,7 +1441,7 @@ mod tests {
         let text = "<think>thought</think>response";
         let mut ctx = ContextState::new();
         ctx.push_no_log(Section::Conversation, AstNode::branch(Role::Assistant, vec![]));
-        let mut p = ResponseParser::new(0);
+        let mut p = ResponseParser::new(0, false);
         for ch in text.chars() {
             p.feed_token(&ch.to_string(), &mut ctx);
         }
@@ -1449,7 +1457,7 @@ mod tests {
         let text = "text<tool_call>\n<function=bash>\n<parameter=command>ls</parameter>\n</function>\n</tool_call>more";
         let mut ctx = ContextState::new();
         ctx.push_no_log(Section::Conversation, AstNode::branch(Role::Assistant, vec![]));
-        let mut p = ResponseParser::new(0);
+        let mut p = ResponseParser::new(0, false);
         let mut tool_calls = 0;
         for ch in text.chars() {
             tool_calls += p.feed_token(&ch.to_string(), &mut ctx).len();
@@ -1497,8 +1505,10 @@ mod tests {
             AstNode::thinking("hmm"),
             AstNode::content("answer"),
         ]);
-        // Thinking renders as empty, content renders as-is
+        // Thinking renders wrapped in <think>...</think> so the model sees
-        assert_eq!(node.render(), "<|im_start|>assistant\nanswer<|im_end|>\n");
+        // previous turns' reasoning (Qwen 3.6 style: CoT stays in the
+        // conversation across turns).
+        assert_eq!(node.render(), "<|im_start|>assistant\n<think>\nhmm\n</think>\nanswer<|im_end|>\n");
     }
 
     #[test]
@@ -1577,10 +1587,19 @@ mod tests {
     fn test_tokenize_invisible_nodes_are_zero() {
         if !init_tokenizer() { return; }
 
-        assert_eq!(AstNode::thinking("deep thoughts").tokens(), 0);
         assert_eq!(AstNode::log("debug info").tokens(), 0);
     }
 
+    #[test]
+    fn test_tokenize_thinking_matches_rendered_tags() {
+        if !init_tokenizer() { return; }
+
+        // Thinking is now prompt-visible (wrapped in <think>...</think>);
+        // token count must match the rendered wrapping.
+        let node = AstNode::thinking("deep thoughts");
+        assert_eq!(node.tokens(), tokenizer::encode(&node.render()).len());
+    }
+
     #[test]
     fn test_tokenize_decode_roundtrip() {
         if !init_tokenizer() { return; }