tools/epro2/std: rewrite to Option 2 (objects dump) per downstream spec

Downstream came back with concrete requirements: don't pre-compute Std
shape[] tilde strings, just dump the raw EPRO2 `objects: {id: payload}`
dict and they'll write a ~100-LoC adapter on their side. Pulling the
tilde-mapping work back saves us from second-guessing positional fields
without their parser to verify against, and shortens our pcb_writer
from ~500 lines to ~40.

Output shape (Std envelope intact, just no `shape[]`):

    {
      "success": true, "code": 0,
      "result": {
        "uuid", "puuid", "title",
        "docType": 3 | 1,
        "components": {},
        "dataStr": {
          "head": {
            "docType": "3" | "1",
            "editorVersion": "facere-epro2/0.1 (epro2 <X.Y.Z>)",
            "units": "mil",
            "epro2_doc_uuid": ...,
            "epro2_editor_version": ...,
          },
          "BBox": {x, y, width, height},   # mil
          "layers": [...],                  # Std layer-string array
          "objects": dict(doc.objects),     # raw EPRO2, 1:1
          "preference": {}, "netColors": [], "DRCRULE": {},
        }
      }
    }

Per-doc spec downstream gave us:
  - shape[] dropped (empty placeholder misleads adapter)
  - all units mil (no mm conversion — Std canvas already declares mil)
  - head.units="mil" so adapter doesn't have to guess
  - BBox min/max across known x/y/startX/endX/centerX fields; adapter
    can refine by walking path arrays itself
  - layers[] keeps Std's 17-line default + inner SIGNAL layers actually
    used (21~Inner1.., 22~Inner2..)
  - empty stubs preference/netColors/DRCRULE for grep-based triage

New: docs/sources/epro2_to_std_mapping.md with the full EPRO2 OPTYPE →
Std verb table that downstream's adapter authors will copy from. Tables
include the layer-id remapping (the 5↔7 paste/mask flip, 11→10 outline,
12→11 multi, SIGNAL 15+→21+), PCB op mappings, SCH op mappings (marked
best-effort: no Std SCH samples in our corpus), and the 5-Voltage
placeholder COMPONENT → extra net flag trick. Extracted from the
previous Option-3 writer (commit fe6971f) so adapter writers don't
have to reverse-engineer it from source.

ESP-VoCat smoke: 6 PCB + 9 SCH = 15 JSON files, head.units=mil
preserved, no shape[] field present. 82 → 84 unit tests pass.

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

log.md

@@ -4,6 +4,70 @@
 
 ---
 
+## 2026-04-29 04:30  std/ writer 翻 Option 2：raw objects dump + mapping doc
+
+**Claude 会话**
+
+接 `fe6971f`。下游同学回了具体规格：选 Option 2（objects dict 直接 dump），不要我们做 tilde 串映射，他自己写 ~100 LoC adapter 翻。把之前 Option 3（full Std `shape: ["TRACK~...", ...]`）那套删了，重写。
+
+### 下游确认的规格
+
+1. **`shape[]` 不要保留**——adapter 从 `objects` 重建，空占位反而误导
+2. **全 mil**——不转 mm，BBox 也 mil（Std `canvas` 那一行 `~mil~` 已经写了）
+3. **`head` 必带**：
+   - `editorVersion` ← `facere-epro2/0.1 (epro2 <doc.head.editVersion>)`，Wokwi 据此选解析分支
+   - `docType` ← `"3"` (PCB) / `"1"` (SCH)
+   - `units` ← `"mil"`
+4. **`layers[]`** 沿用 Std 53-layer 字符串数组格式（`"1~TopLayer~#FF0000~..."`），inner SIGNAL 有用到才追加
+5. **保留空 stub** `preference / netColors / DRCRULE`——失败时 grep 路径稳定
+6. **per-doc 一个 .json**，平铺，不合并
+
+### 写完三件
+
+#### `tools/epro2/std/pcb_writer.py` 重写
+40 行核心逻辑（之前 Option 3 是 500 行）：
+- `_gather_bbox_points`：扫所有 op 的 `x/y/startX/...` 等坐标字段，min/max
+- `_used_inner_signal_layers`：找真正有 primitive 的 SIGNAL 内层 id
+- envelope 直接 dump `dict(doc.objects)`
+
+#### `tools/epro2/std/sch_writer.py` 重写
+更短——schematic 没 copper layer，layers=[]。其它结构跟 PCB 一样。
+
+#### `docs/sources/epro2_to_std_mapping.md` 新增
+**这是给下游 adapter 的关键文档**——他写 100 LoC adapter 时按这个表查。内容：
+- EPRO2 layer id → Std layer id 重映射表（最坑：5↔7 mask/paste 反着、11→10 outline、12→11 multi、SIGNAL 15+→Std 21+）
+- PCB OPTYPE → Std verb 全表（TRACK / VIA / COPPERAREA / SOLIDREGION / CIRCLE / LIB+#@$PAD+TEXT）
+- SCH OPTYPE → Std verb 全表（W / N / T / LIB+#@$P）—— 标了 best-effort，没 Std SCH 实样
+- COMPONENT placement 旋转 + 平移公式（footprint-local → PCB-absolute）
+- 5-Voltage 占位符 `pid8a0e77bacb214e` 的 Global Net Name → 额外 N flag 的 trick
+
+mapping doc 直接从 `fe6971f` 那个 Option 3 writer 提炼出来——他不用读我们代码，照表填就行。
+
+### 实测
+
+ESP-VoCat 6 PCB + 9 SCH = 15 JSON：
+- 一个典型 PCB: `objects=1719, layers=17, BBox=(-2293, -900, 2997×1899)`
+- `head.units = "mil"`、`head.editorVersion = "facere-epro2/0.1 (epro2 3.2.91)"`、`head.docType = "3"`
+- `shape` 字段已确认**不存在**
+- `objects` 原 payload 1:1 保留（含 LAYER ops 的 activeColor 等所有字段）
+
+### 决策（Why）
+
+- **不留空 `shape: []`**——下游说"误导 adapter"。明确不存在，比空数组更诚实
+- **`head.editorVersion` 加前缀 `facere-epro2/0.1`**——区分我们的输出 vs lceda 真实 Std；adapter 看到这个能猜出是 EPRO2 转过来的
+- **保留 `preference/netColors/DRCRULE` 空 stub**——下游说失败 grep 排查方便
+- **mapping doc 单独成文不混在 README**——adapter 作者一个文件就够，不用读源码
+
+### 测试
+
+82 → 84 单测全过：原 Option 3 的 11 个测试改成验 Option 2（envelope 必带 key / 无 shape / head units&version / objects 1:1 / BBox min-max / 内层 SIGNAL append / docType reject）。
+
+### Push
+
+`fe6971f` 的 Option 3 已被这次 commit 覆盖 / 简化掉。下游回来如果说 mapping table 有错位再修——但他自己拿表填 adapter，跑通后给反馈我们再迭代。
+
+---
+
 ## 2026-04-29 04:00  Std-format JSON 转换器：EPRO2 → 下游同学 Wokwi pipeline 的输入格式
 
 **Claude 会话**