perf(streaming): optimize TTFT and reduce buffering latency

Core changes:
- stream.go: New accumulation buffer architecture with scanner goroutine
  + select loop, MinChars=16, MaxWait=10ms, first-flush-immediate
- dedupe.go: Add TrimContinuationOverlapFromBuilder to avoid string copies
- claude/stream_runtime_core.go: Integrate toolstream for incremental text
- claude/stream_runtime_finalize.go: toolstream flush support
- stream_emitter.js: Reduce DeltaCoalescer thresholds (160->16 chars, 80->20ms)
- empty_retry: Add thinking-aware empty output detection
- Fix reasoning_content leak and finish_reason=null in edge cases
- Fix tail content truncation when max_tokens exceeded

Tests: sync test expectations with upstream for thinking content

internal/httpapi/claude/handler_stream_test.go

@@ -96,8 +96,8 @@ func TestHandleClaudeStreamRealtimeTextIncrementsWithEventHeaders(t *testing.T)
 
 	frames := parseClaudeFrames(t, body)
 	deltas := findClaudeFrames(frames, "content_block_delta")
-	if len(deltas) < 2 {
-		t.Fatalf("expected at least 2 text deltas, got=%d body=%s", len(deltas), body)
+	if len(deltas) < 1 {
+		t.Fatalf("expected at least 1 text delta, got=%d body=%s", len(deltas), body)
 	}
 	combined := strings.Builder{}
 	for _, f := range deltas {