docs: improve readability of PROTEAN comparison figures

docs/research/ORIGINAL_VS_PROPOSED_DEEP_COMPARISON.md

@@ -133,7 +133,9 @@ classes while giving up a little rare-class coverage.
 
 ## 6. Paired per-client results (test, vs original_protean)
 
-Figure I: `artifacts/figures/final/figI_per_client_original_vs_proposed.png`.
+Figures I1/I2 (one per dataset):
+`artifacts/figures/final/figI1_xiiotid_per_client_original_vs_proposed.png`,
+`artifacts/figures/final/figI2_5gnidd_per_client_original_vs_proposed.png`.
 Per-observation table: `deepcmp__per_client.csv` (30 observations per condition).
 
 | Condition | beta* | improved | degraded | unchanged | mean Δ | min Δ | max Δ |
@@ -159,7 +161,9 @@ Counts use the ±0.02 effectively-unchanged band on per-client test rare-class m
 
 ## 7. Per-class precision/recall/F1 changes (beta>0 conditions)
 
-Figure J: `artifacts/figures/final/figJ_per_class_original_vs_proposed.png`.
+Figures J1/J2 (one per dataset):
+`artifacts/figures/final/figJ1_xiiotid_per_class_original_vs_proposed.png`,
+`artifacts/figures/final/figJ2_5gnidd_per_class_original_vs_proposed.png`.
 Source: `deepcmp__per_class.csv` (per seed) aggregated as the mean delta over the three seeds; class
 support is the mean test support over seeds. X-IIoTID and 5G-NIDD labels are never mixed.
 
@@ -198,8 +202,11 @@ Points to note:
 
 ## 8. Confusion-matrix difference analysis
 
-Figure K shows the paired difference for **5G-NIDD, alpha=0.75, seed=0** (the same pair as Figure H):
-`artifacts/figures/final/figK_confusion_delta_5gnidd_a0.75_seed0.png`. Rows = true class, columns =
+The ranked-change figure (primary) is
+`artifacts/figures/final/figK_confusion_error_changes_5gnidd_a0.75_seed0.png`; the full difference
+matrices (raw count and row-rate) are kept as the supplemental
+`artifacts/figures/final/figK_confusion_delta_5gnidd_a0.75_seed0.png`. Both describe the paired
+difference for **5G-NIDD, alpha=0.75, seed=0** (the same pair as Figure H). Rows = true class, columns =
 predicted class; both matrices have identical class order and 243,177 test samples. Source:
 `deepcmp__confusion_delta.csv`.
 
@@ -227,8 +234,8 @@ Largest raw-count changes for this pair (proposed − original):
   are X-IIoTID `Reconnaissance → Exploitation` (−6,773), `Reconnaissance → Weaponization` (−3,398),
   `Lateral_movement → Tampering` (−1,699) and 5G-NIDD `Benign → TCPConnectScan` (−1,484); the largest
   **increased** ones are X-IIoTID `Normal → C&C` (+2,546) and `Weaponization → Exploitation` (+1,286).
-  Raw counts are support-sensitive, so these are descriptive; the row-normalized view (Figure K,
-  right panel) is the support-fair comparison.
+  Raw counts are support-sensitive, so these are descriptive; the row-normalized view in the
+  supplemental Figure K matrix (right panel) is the support-fair comparison.
 
 ## 9. Support imbalance versus client outcomes