feat: capture thinking text from MITM-intercepted API responses

The LS strips thinking/reasoning text from plannerResponse steps —
only the thinkingSignature (opaque verification blob) is preserved.
The actual thinking text flows through the MITM proxy in the raw
Google SSE response (parts with thought: true) and Anthropic SSE
(thinking_delta content blocks).

Changes:
- StreamingAccumulator now accumulates thinking text from SSE events
- ApiUsage gains thinking_text: Option<String>
- usage_from_poll returns (Usage, Option<thinking_text>)
- Thinking text priority: MITM-captured > LS-extracted (fallback)
- Reasoning output item now populated from real API data
- Removed debug dump code

src/mitm/store.rs

@@ -22,6 +22,10 @@ pub struct ApiUsage {
     pub cache_read_input_tokens: u64,
     /// Google-specific: thinking/reasoning output tokens (extended thinking)
     pub thinking_output_tokens: u64,
+    /// The actual thinking/reasoning text from the model.
+    /// Captured from Google SSE parts with `thought: true` or Anthropic thinking blocks.
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub thinking_text: Option<String>,
     /// Google-specific: response output tokens (non-thinking portion)
     pub response_output_tokens: u64,