Add missing reference to Figure 3 (progressive disclosure timeline)

Figure 3 (progressive-timeline) was created but never referenced in the text,
leaving readers without guidance on when to consult it.

Added reference at line 627 in the Pattern Implementation subsection, right
after introducing the dormant/activation concept via code listings. The
reference reads: 'Figure 3 visualizes this activation timeline across the
curriculum.'

This ensures all figures in the paper are properly referenced and integrated
into the narrative flow. All other figures and tables were already correctly
referenced.

Reference audit:
✓ Figure 1 (code-comparison) - line 183
✓ Figure 2 (module-flow) - line 290
✓ Figure 3 (progressive-timeline) - line 627 [NEW]
✓ Table 1 (framework-comparison) - line 421
✓ Table 2 (objectives) - line 478
✓ Table 3 (performance) - lines 811, 1013

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

paper/paper.tex

@@ -624,7 +624,7 @@ Traditional ML education faces a pedagogical dilemma: students need to understan
 
 \subsection{Pattern Implementation}
 
-TinyTorch's \texttt{Tensor} class includes gradient-related attributes from Module 01, but they remain dormant until Module 05 activates them through monkey-patching (\Cref{lst:dormant-tensor,lst:activation}).
+TinyTorch's \texttt{Tensor} class includes gradient-related attributes from Module 01, but they remain dormant until Module 05 activates them through monkey-patching (\Cref{lst:dormant-tensor,lst:activation}). \Cref{fig:progressive-timeline} visualizes this activation timeline across the curriculum.
 
 \begin{lstlisting}[caption={Module 01: Dormant gradient features.},label=lst:dormant-tensor,float=t]
 # Module 01: Foundation Tensor