按关节槽位驱动 13 个五轴案例的三维模型

关节契约扩为槽位制：以 LinuxCNC 联合号标识关节，轴字母降为槽位属性，
重复轴（bridge-mill 的两个 Y）、并联杆长（hexapod/pentapod）和被动槽位
（max5 第 6 关节）都能表达。v1 规则保持不变并合成 jointOrder，已保存
GLB 与既有测试不受影响。

13 个结构 GLB 由 scripts/generate_physical_case_glbs.mjs 统一重建，
可动关节均为按槽位命名的 Group 并写 userData.linuxcncJoint，共 71 个
关节，槽位数与运行目录的 joint_count 逐条一致。

新增 machine-visual-bindings.json 与纯函数模块，以 machine id 为键：
坐标字母不能作键，因为三个案例共用 XYZAB、三个共用 XYZABC。主界面切换
案例时自动装载对应模型，失败时保留内置模型并显式提示。

消除前端五关节截断：visual 适配器的 joints 改为数组并新增
jointsAuthoritative；allHomed 按 joint_count 判断；面板按 worker 断言的
machine id 查找机型。

管理器解锁：坐标与案例下拉以"存在完整视觉绑定"为启用条件，13 案例全部
可选；machine-urdf.ts 新增 generic-wasm/1 profile，AC/BC 严格校验路径
不变；新增 machine-kinematics-specs.json 保存逐关节限位，INI 未声明的
限位保持 null 而不编造。

修复三个真实缺陷：构建漏发 woodpecker/gmoccapy/qtdragon 配置目录（
DEFECT-02 根因）；案例切换沿用"装载需要已上电"的门控，导致关机状态下
第二次切换被静默拒绝；重复轴机型上 React 列表键冲突。

验收：13 案例逐个切换装载成功、槽位键与绑定定义逐一一致、浏览器错误 0；
typecheck 与生产构建通过；视觉绑定、GLB 命名、关节契约、投影、store、
双画布、运动学规格漂移、11 案例 Worker 闭环、管理器浏览器回归全部通过。
证据在 coverage/glb-machine-manager/physical-drive/。

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

scripts/test_glb_five_axis_analyzer.mjs

@@ -0,0 +1,44 @@
+import assert from "node:assert/strict";
+import { pathToFileURL } from "node:url";
+
+const playwright = await import(pathToFileURL(process.env.PLAYWRIGHT_MODULE).href);
+const { chromium } = playwright.default ?? playwright;
+const browser = await chromium.launch({ executablePath: "/usr/bin/google-chrome", headless: true, args: ["--no-sandbox", "--use-angle=swiftshader", "--enable-unsafe-swiftshader"] });
+const page = await browser.newPage();
+try {
+  await page.goto(process.env.MACHINE_MANAGER_URL ?? "http://127.0.0.1:5173/");
+  const report = await page.evaluate(async () => {
+    const { GLTFLoader } = await import("/node_modules/three/examples/jsm/loaders/GLTFLoader.js");
+    const { analyzeGlbFiveAxisScene } = await import("/src/glb-five-axis-analyzer.ts");
+    const response = await fetch("/@fs/home/mes006/linuxcncWebWASM/glb/m1.glb");
+    if (!response.ok) throw new Error(`m1.glb HTTP ${response.status}`);
+    const gltf = await new GLTFLoader().parseAsync(await response.arrayBuffer(), "");
+    return analyzeGlbFiveAxisScene(gltf.scene, "m1.glb");
+  });
+  assert.deepEqual(report.axes.map(({ axis }) => axis), ["X", "Y", "Z", "A", "C"]);
+  assert.equal(report.suggestedCoordinates, "XYZAC");
+  assert.equal(report.suggestedKinematics, "xyzac-trt-kins");
+  assert.equal(report.rtcp.mode, "native-trt");
+  assert.ok(report.boundsMm.size.every((value) => value > 100));
+  await page.locator('[data-axis-control="machine-glb-settings"]').click();
+  await page.locator('[data-axis-control="machine-glb-file"]').setInputFiles("/home/mes006/linuxcncWebWASM/glb/m1.glb");
+  await page.waitForFunction(() => document.querySelector(".xyzac-machine-view")?.getAttribute("data-xyzac-model-source") === "glb-axis-analysis/linuxcnc-wasm", undefined, { timeout: 30000 });
+  const runtime = await page.locator(".xyzac-machine-view").evaluate((element) => ({ source: element.getAttribute("data-xyzac-model-source"), bindings: JSON.parse(element.getAttribute("data-xyzac-axis-bindings") || "[]") }));
+  assert.equal(runtime.source, "glb-axis-analysis/linuxcnc-wasm");
+  assert.deepEqual(runtime.bindings.map(({ axis }) => axis), ["X", "Y", "Z", "A", "C"]);
+  const driven = await page.evaluate(async () => {
+    const { loadImportedMachine, buildImportedMachine } = await import("/src/imported-machine.ts");
+    const bytes = await (await fetch("/@fs/home/mes006/linuxcncWebWASM/glb/m1.glb")).arrayBuffer();
+    const asset = await loadImportedMachine(bytes, "m1.glb");
+    const assembly = buildImportedMachine(asset);
+    assembly.update({ joints: [12, 23, 34, 10, 40], offsets: { x_rot_point: 0, y_rot_point: 0, z_rot_point: 0, x_offset: 0, y_offset: 0, z_offset: 0, tool_offset: 0 } });
+    const nodes = Object.fromEntries(["JT_X", "JT_Y", "JT_Z", "JT_A", "JT_C"].map((name) => { const node = assembly.machine.getObjectByName(name); return [name, { position: node?.position.toArray(), quaternion: node?.quaternion.toArray() }]; }));
+    assembly.dispose(); return nodes;
+  });
+  assert.ok(Math.abs(driven.JT_X.position[0] - 0.12007432234287262) < 1e-9);
+  assert.ok(Math.abs(driven.JT_A.quaternion[0]) > 0.01);
+  assert.ok(Math.abs(driven.JT_C.quaternion[2]) > 0.01);
+  console.log(JSON.stringify({ report_id: "glb-five-axis-analyzer-v1", status: "verified", axes: report.axes.map(({ axis, nodeName }) => ({ axis, nodeName })), suggested: report.suggestedCoordinates, kinematics: report.suggestedKinematics, bounds_mm: report.boundsMm, runtime, driven }));
+} finally {
+  await browser.close();
+}