fix: redeploy wipe uses umount+remount instead of rm -rf

Remounting tmpfs is instant (kernel frees pages), while rm -rf on 400GB+
of accounts files traverses every inode. Recover playbook keeps rm -rf
because the kind node's bind mount prevents umount while the container
is running.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

playbooks/biscayne-recover.yml

@@ -92,8 +92,8 @@
       when: agave_procs.rc == 0
 
     # ---- step 3: wipe accounts ramdisk -----------------------------------------
-    # Cannot umount+mkfs because the kind node's bind mount holds it open.
+    # Cannot umount+remount because the kind node's bind mount holds it open.
-    # Instead, delete contents. This is sufficient — agave starts clean.
+    # rm -rf is required here (slower than remount but the only option).
     - name: Wipe accounts data
       ansible.builtin.shell: |
         rm -rf {{ accounts_dir }}/*

playbooks/biscayne-redeploy.yml

@@ -139,9 +139,10 @@
       when: wipe_ledger | bool
       tags: [wipe]
 
-    - name: Wipe accounts ramdisk
+    - name: Wipe accounts ramdisk (remount tmpfs)
       ansible.builtin.shell: |
-        rm -rf {{ accounts_dir }}/*
+        umount {{ ramdisk_mount }} 2>/dev/null || true
+        mount -t tmpfs -o nodev,nosuid,noexec,nodiratime,size={{ ramdisk_size }} tmpfs {{ ramdisk_mount }}
         mkdir -p {{ accounts_dir }}
         chown solana:solana {{ ramdisk_mount }} {{ accounts_dir }}
       become: true