Run whole-data archives and restores in the background

A data archive or restore ran inside its request, which sent nothing
until tar finished: Bun closes a silent socket after 10s (its default
idleTimeout in production) and a proxy cuts it at about 100s, so the
page showed a network error while the archive kept going, and a retry
said another backup was already running.

The routes now start the run (startDataBackupRun, refused while the
backup lock is held) and answer at once. The page follows it by polling
api/admin/backups/data-status: an archive reloads once written, a
restore once the status comes from a new process after its restart,
and a failure shows its error. A run in progress survives a reload.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

src/data/app-data.ts

@@ -607,6 +607,13 @@ export const AppData = {
      */
     DataBackupNamePrefix: "cobble-data-",
     DataBackupsKept: 5,
+    /**
+     * How often the backups page asks how a whole-data archive or restore
+     * is going, and how long it waits for the app to come back after a
+     * restore restarted it before it reloads anyway.
+     */
+    DataBackupPollMs: 2000,
+    DataRestoreWaitMs: 120000,
     /**
      * Responses smaller than this go out uncompressed: the encoding
      * costs more than it saves.