feat(skills): add dryRun flag for scan prune path

Add a `dryRun` option to the scan-projects endpoint. When true, the
scan identifies which skills would be pruned and which agents would be
affected, but does not delete anything or modify agent configs.

The response now includes:
- `pruned[]`: list of skills that would be (or were) removed, with
  affected agent names
- `dryRun`: boolean echoed back so callers can distinguish preview
  results from live mutations

This lets callers preview destructive prune operations before committing
to them, addressing the review concern about silent deletion of
production data.

Co-Authored-By: Paperclip <noreply@paperclip.ing>

server/src/__tests__/company-skills-routes.test.ts

@@ -78,7 +78,9 @@ describe("company skill mutation permissions", () => {
       updated: [],
       skipped: [],
       conflicts: [],
+      pruned: [],
       warnings: [],
+      dryRun: false,
     });
     mockLogActivity.mockResolvedValue(undefined);
     mockAccessService.canUser.mockResolvedValue(true);
@@ -293,9 +295,13 @@ describe("company skill mutation permissions", () => {
       updated: [],
       skipped: [],
       conflicts: [],
+      pruned: [
+        { skillId: "skill-1", slug: "ghost-skill", key: "vercel-labs/agent-browser/ghost-skill", sourceLocator: "https://github.com/vercel-labs/agent-browser", affectedAgents: ["Builder"] },
+      ],
       warnings: [
         'Skill "ghost-skill" was removed from https://github.com/vercel-labs/agent-browser and detached from Builder.',
       ],
+      dryRun: false,
     });
 
     const res = await request(await createApp({