feat(analyzer): surface TTFT/TPOT as features and CDF figures

The gateway records three epoch-ms stamps per request (admitted, forwarded to
backend, backend first token). They reached *-raw.jsonl but stopped there:
features.csv had only total latency, and none of the figures covered
time-to-first-token.

Derive ttft_ms, ttft_e2e_ms and tpot_ms in one helper shared by the streaming
path (preparation.py) and the in-memory path (features.py) so the two cannot
drift, and add figures 14 (TTFT, backend vs end-to-end) and 15 (TPOT). The two
TTFT series are plotted together because their offset is the gateway admission
delay, which is then readable straight off the chart.

Rows missing a backend stamp and single-token responses are recorded as 0 and
filtered at plot time; left in, they would stack at the origin and pull every
percentile down. Figure 15 is a single panel: TPOT is near-deterministic
(p50 21.9, p90 23.0), so a p90 zoom renders a flat line and a vertical edge.

Verified by recomputing ttft_ms straight from the original SLS rows for all
2190 requests of a sample: 0 mismatches.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

README.md

@@ -10,7 +10,7 @@
 - `trace_analyzer` 不再直接处理 provider-native 原始日志。
 - `details/` 只放画图所需数据。
 - `figures/` 只放最终图片。
-- 当前主分析入口只维护 13 张图，对应的数据也只保留为这 13 张图服务的最小集合。
+- 当前主分析入口只维护 15 张图，对应的数据也只保留为这 15 张图服务的最小集合。
 
 朴素 wall-clock 时间字段如 `time` 按 `UTC+8` 解释。
 
@@ -113,6 +113,7 @@ outputs/
 
 - `request_metrics.csv`
   请求级主表。用于：
+  - TTFT / TPOT
   - input/output length
   - session turns / turn 维度请求长度
   - trigger role 占比
@@ -131,7 +132,7 @@ outputs/
 - `progress.json`
   大 trace 运行时的进度与资源占用信息。
 
-### `figures/` 当前只保留的 13 张图
+### `figures/` 当前只保留的 15 张图
 
 `figures/` 是展示层，只放图片和轻量说明文件。
 
@@ -148,6 +149,8 @@ outputs/
 11. `11_alive_kvcache_blocks_timeline.png`
 12. `12_bucket_kvcache_reuse_ratio.png`
 13. `13_session_cross_bucket_kvcache_miss.png`
+14. `14_ttft_cdf.png`
+15. `15_tpot_cdf.png`
 
 另外还有：