feat: add direct edit handles behind directEdit flag

Steps 1-3 of the main map road interval editor.

EditSession keeps the command stack, undo/redo and previewSeq arbitration as
pure logic with no OpenLayers reference, so all of it is unit-tested in node.
Pointer displacement converts to meters through EPSG:4326 and spherical
distance: treating a 3857 delta as meters desyncs the geometry from the cursor
by 1/cos(latitude). Handle drags project onto the axis the manifest declares
and clamp to its range, so the client never writes a coordinate into a road
polygon.

All of it sits behind a directEdit flag that defaults to off. With the flag off
the workbench requests no manifest, creates no extra source and registers no
interaction, so behaviour matches main.

The ol-ext probe passed its three gates but is not adopted for road handles.
Transform translates by the raw pointer delta, so a handle detaches from its
clamped constraint value: a drag reading -24.1 m produced a draft of -5.4 m.
Production needs the handle position derived from the constraint instead, which
means owning the position update, so native OL PointerInteraction will carry
the drag. ol-ext stays out of package.json; the probe is kept as a manual
harness. Reserve handles are unreachable with the current solver, recorded in
research/ rather than worked around.

Also names the dead backend when an API response is empty, instead of
surfacing "Unexpected end of JSON input" from response.json().

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

.trellis/tasks/08-26-direct-edit-map-editor/research/reserve-handles-unreachable.md

@@ -0,0 +1,70 @@
+# reserve 内手柄在当前服务端实现下不可达
+
+第 3 步人工验证时发现：点遍所有道路都看不到灰色（不可拖）手柄。经实测确认这是**正确行为**，不是漏测。
+
+## 实测数据
+
+`test/fixtures/fengshu-er-road.osm`，24 条方向道路、32 个 reserve：
+
+```
+road handles        : 126
+road NOT editable   : 0
+with disabledReason : 0
+junction handles    : 96   （JunctionTools 所有，主地图已过滤）
+```
+
+## 为什么不可达
+
+`makeRoadHandles()` 里 `editable: false` 的唯一来源是 `unavailable`：
+
+```js
+const start = Math.min(1, startReserve);
+const end = Math.max(0, 1 - endReserve);
+const unavailable = start >= end;
+```
+
+而 `junctionReserves()` 给每端的保留区比例有硬上限：
+
+```js
+const fraction = Math.min(0.45, cutback / length);
+```
+
+于是 `startReserve ≤ 0.45`、`endReserve ≤ 0.45`，即 `start ≤ 0.45` 且 `end ≥ 0.55`。
+`start >= end` 结构上永远为假，`unavailable` 分支是死代码。
+
+实测输出里能直接看到上限生效的样子——两端都撞到 0.45 的短路段：
+
+```
+segment:way/858770822/1   window=[0.450,0.550]
+segment:way/117947564/0   window=[0.450,0.550]
+segment:way/851989492/2   window=[0.450,0.550]
+```
+
+另外手柄的 station 取自 `(start + end) / 2`，即无保留区窗口的正中，
+所以三类道路手柄的位置也永远不会落进 reserve。
+
+## 对验收标准的影响
+
+prd.md 的「落在 junction reserve 内的手柄不可拖动，并提示进入 JunctionTools」
+对当前三类道路手柄是**空真**：它们永远不落在 reserve 内，所以约束自动成立。
+
+客户端一侧的置灰与原因提示路径有单测覆盖（`selection.test.ts`：disabled 手柄被保留、
+`disabledReasonOf` 始终给出原因），只是服务端从不产生这种输入。因此第 3 步不需要改动。
+
+reserve 边界真正变得用户可见是在第 4 步的**区间范围手柄**：它可以被拖向 reserve 边界，
+由 `projectIntervalEnd()` 钳到 `window.minStation` / `maxStation`。届时这条边界才有可演示的行为。
+
+## 留给用户决定的设计问题
+
+那个死分支暗示原作者期望「短路段应完全归 JunctionTools」。但 0.45 的上限把结果改成了：
+一条几何上被两个大路口主导的短路，仍然会在正中间得到一条仅占全长 10% 的可编辑带
+（上面 `[0.450,0.550]` 那几条）。
+
+两种取向都讲得通，需要产品判断：
+
+- 保持现状：任何道路都留一条可编辑带，哪怕很窄。
+- 改为：当 `cutback / length` 两端之和超过某阈值时，整条路标 `editable: false`，
+  引导用户进入 JunctionTools。这会让死分支复活，也让验收标准变成可演示的。
+
+改动落在 `src/compile/direct-edit-solver.js`（属已归档的 `direct-edit-solver-api` 任务范围），
+不在本任务范围内，故此处只记录不实施。