fix(skills): stop asserting readdir order in the skill root walk test

The Windows skill-sharing release gate — which blocks publish-release —
failed on this test, wedging the 1.4.198 cut. The implementation is
correct; the assertion was not.

`findSkillFiles` pushes results in `readdir` order and dedupes visited
directories by realpath, so of the 32 junctions pointing at one target
exactly one survives. The assertion hardcoded both the array order and
which link won. `readdir` order is filesystem-dependent: APFS and ext4
return SKILL.md first, while NTFS enumerates its filename index
alphabetically on the uppercased name, where "LINK00" sorts before
"SKILL.MD" (L=0x4C < S=0x53). Windows therefore returned the same two
paths in the opposite order.

Assert the contract instead: the real file is present, exactly one
link-routed path survives dedup, and nothing else does.

(cherry picked from commit 9a832e9bda)
(cherry picked from commit 505b85cd0d)

src/main/skills/skill-root-file-walk.test.ts

@@ -66,10 +66,17 @@ describe('findSkillFiles', () => {
 
     expect(await findSkillFiles(root, 4)).toEqual([join(edge, 'SKILL.md')])
     expect(statPaths).toEqual([])
-    expect(await findSkillFiles(root, 5)).toEqual([
-      join(edge, 'SKILL.md'),
-      join(edge, 'link00', 'SKILL.md')
-    ])
+    // Why not a fixed array: `readdir` order is filesystem-dependent, and both
+    // the result order and which link survives dedup follow it. NTFS enumerates
+    // its name index alphabetically, so `link00` precedes `SKILL.md` on Windows
+    // and follows it on APFS/ext4. All 32 links share one realpath, so the
+    // visited set collapses them to a single entry beside the real file.
+    const withinDepth = await findSkillFiles(root, 5)
+    expect(withinDepth).toContain(join(edge, 'SKILL.md'))
+    expect(withinDepth.filter((path) => /[\\/]link\d{2}[\\/]SKILL\.md$/.test(path))).toHaveLength(
+      1
+    )
+    expect(withinDepth).toHaveLength(2)
     expect(statPaths).toHaveLength(32)
   })