docs(kvc): replace v2 path breakdown tables with generated figures

V2_DEEP_ANALYSIS §3.1 (execution_mode distribution) and §3.2 (path-level
latency vs DP) had hand-typed tables with approximate latencies (e.g.
"~1.0s") and required readers to mentally compare 5+ rows × 5 columns.
Both sections now reference generated PNG figures derived directly from
the v2 + DP metrics.jsonl files.

§3.1 figure (v2_execution_mode_distribution.png):
  Horizontal bar chart, log x-axis. 4076 direct-to-D fast-path requests
  (green) dwarf the rest by ~30x; the long tail of slow / fallback /
  failure modes is visible at one glance. Counts and percentages
  annotated on each bar.

§3.2 figure (v2_path_level_latency.png):
  Grouped bar chart, log y-axis. Per-path TTFT p50 / TTFT p99 / Lat p50
  with exact numeric labels (no more "~1.0s" approximations). Sample
  counts annotated below each path. Quick visual reads:
   - KVC fast path TTFT p50 41ms vs DP 92ms (2.2x faster)
   - KVC reseed TTFT p99 5.12s vs DP 0.43s (12x slower) -- the cost
   - KVC no-d-capacity TTFT p99 7.65s (worst case)

Bundled:
- scripts/analysis/plot_v2_path_breakdown.py -- the script that
  generates both figures; rerunable when v2 data changes.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

docs/V2_DEEP_ANALYSIS_ZH.md

@@ -187,31 +187,22 @@ v2 整体跑得快不仅因为 "KVC 机制好"，更因为 **91.6% 请求被路
 
 ### 3.1 v2 内部 execution_mode 分布
 
-```
-kvcache-direct-to-d-session              4076  (91.6%)  ← 快路径
-pd-router-d-session-reseed                150  (3.4%)   ← 慢路径：mooncake reseed
-pd-router-fallback-session-not-resident-session-cap        50  (1.1%)
-pd-router-fallback-session-not-resident-seed-filter-early-turn  52  (1.2%)
-pd-router-turn1-seed                       52  (1.2%)
-pd-router-fallback-no-d-capacity            31  (0.7%)
-pd-router-fallback-real-large-append-session-cap  25  (0.6%)
-其余                                      ~13  (<0.3%)
-```
+![KVC v2 execution_mode 分布](figures/v2_execution_mode_distribution.png)
+
+数据来源：`outputs/qwen3-30b-tp1-ts1-migration-v2/kvc_1p3d_migration_v2_run1_metrics.jsonl`，n = 4449（全部请求，含失败）。绿色 = direct-to-D 快路径 = 91.6%；其余红色 = 慢路径 / fallback / 失败。绘图脚本：`scripts/analysis/plot_v2_path_breakdown.py`。
 
 ### 3.2 path-level 延迟 vs DP
 
-| Path | n | 占比 | TTFT P50 | TTFT P99 | Lat P50 |
-|---|---:|---:|---:|---:|---:|
-| KVC `direct-to-d-session` | 4076 | 91.6% | **42ms** | 89ms | 0.470s |
-| KVC `d-session-reseed` | 150 | 3.4% | ~1.0s | 7.51s | ~2.5s |
-| KVC `no-d-capacity` fallback | 31 | 0.7% | ~1.0s | 7.66s | ~3.0s |
-| KVC `session-not-resident-*` | 102 | 2.3% | 0.3-2.3s | – | 1-3s |
-| DP `dp-colo-router` | 4449 | 100% | 90ms | **427ms** | 0.659s |
+![Path-level latency: KVC v2 各路径 vs DP](figures/v2_path_level_latency.png)
+
+数据来源：同上 + `outputs/qwen3-30b-tp1-ts1-validation/dp4_metrics.jsonl`。Y 轴 log 刻度（latency 跨度 41ms ~ 7.71s）。已过滤 abort / error 请求，所有数字按对等口径计算。
 
 **关键事实**：
-- KVC 的 91.6% fast path 在 TTFT 上确实碾压 DP（42ms vs 90ms）
-- 但 KVC 的 8.4% slow path 比 DP 慢 7-17 倍（1-7s vs 0.4s）
-- DP **没有 slow path**——每个请求在本地 worker 完整 prefill，最坏情况也是 P99 8s 量级
+- KVC 的 91.6% **fast path** 在 TTFT p50 上是 **41ms vs DP 92ms**——压制 DP 2.2×；TTFT p99 150ms vs DP 428ms 仍优 2.9×
+- KVC 的 **3.4% reseed 慢路径** TTFT p99 = **5.12s**，是 DP 单一路径 p99（428ms）的 **12×**
+- KVC 的 **0.7% no-d-capacity fallback** 是最坏情况：TTFT p99 = 7.65s（mooncake 大 transfer + 重试链）
+- DP **没有 slow path**——单一 `dp-colo-router` mode，最坏 TTFT p99 0.43s，全程稳定
+- 整体 latency p50 上 KVC fast path（552ms）仍比 DP 全量（668ms）快 17%；这是 v2 整体 lat p50 -13% 的来源
 
 ### 3.3 Fast path 的工作量比 DP 少 6.9× —— 不是 mechanism 更快