[PyTokMHD] Tearing mode shows decay despite q=2 rational surface

[callme-YZ]

Problem

Tearing mode IC satisfies all theoretical requirements but shows decay (γ<0) instead of growth.

Current Status

IC has all requirements:

• ✅ Current sheet (J reversal)
• ✅ Rational surface q≈2 at r≈0.71
• ✅ m=2 perturbation (86% of total)

But simulation shows:

• ❌ Decay: γ ≈ -0.54
• ❌ Growth factor 0.989 over 200 steps

Hypotheses

1. Grid too coarse (32×64)?
2. Perturbation not unstable eigenmode?
3. Numerical diffusion dominates?
4. Missing 3D physics?

Diagnostic Work Done

• IC design fixed (added q_correction)
• Level 1 verification complete
• Tested η=0.0 to 0.05 (all decay)

Next Steps

1. Grid convergence test
2. Eigenmode calculation
3. Benchmark vs BOUT++/M3D-C1

Impact

• Blocks: Tearing mode RL
• Workaround: Use ballooning for v3.0

Priority: P2 (defer to v3.1)

Author: 小P ⚛️
Date: 2026-03-25
Time: ~2h diagnostic