Go on with a backup when the destination marker cannot be checked (closes #389)

A source host whose credentials allow only PUT, which the documents
say is enough to back up, failed every backup after its first: the
marker check looks the marker up with an S3 HEAD, which those
credentials are refused, and any error other than not found was
fatal.

Such an error now means the check could not be made: it is logged at
info level, so --cron stays silent, and the command goes on. Not found
still stops it. The documents now say so, and that such a host does
not notice an emptied destination, so vaultik database delete must be
run on it after the destination is emptied. The fault-injecting test
store gains a Stat hook for the test.

Model: opus-5-5

ARCHITECTURE.md

@@ -312,7 +312,7 @@ Repository interfaces:
 ```
 CreateSnapshot(opts)
     │
-    ├─► markDestination()   // Refuse a destination without the marker, then write it
+    ├─► markDestination()   // Check the marker (see Storage Layout in S3), then write it
     │
     ├─► PruneDatabase()   // Critical: avoid dedup errors
     │
@@ -382,15 +382,20 @@ bucket/
         └── manifest.json.zst        # Blob list (for pruning/verification)
 ```
 
-A destination without `vaultik-destination.txt` is treated as not present
-while the local index holds snapshots stored there (`checkDestinationMarker`
-in `internal/vaultik/destination_marker.go`): `snapshot create` refuses to
-write to it, `prune`, `snapshot purge` and `remote nuke` fail, and
-`snapshot list` and `snapshot remove` behave as for a destination that cannot
-be listed. This keeps those commands from taking an unplugged volume's empty
-mount point for a destination whose snapshots are all gone. The `file://`
-backend never creates its destination directory, so a write to a missing one
-fails.
+A destination that answers a look-up of `vaultik-destination.txt` with not
+found is treated as not present while the local index holds snapshots stored
+there (`checkDestinationMarker` in `internal/vaultik/destination_marker.go`):
+`snapshot create` refuses to write to it, `prune`, `snapshot purge` and
+`remote nuke` fail, and `snapshot list` and `snapshot remove` behave as for a
+destination that cannot be listed. This keeps those commands from taking an
+unplugged volume's empty mount point for a destination whose snapshots are
+all gone. When the look-up fails for any reason other than not found, for
+example when S3 refuses it to credentials that allow only PUT, the check is
+skipped, so a host with such credentials does not notice an emptied
+destination; see
+[docs/REPOSTRUCTURE.md](docs/REPOSTRUCTURE.md#marker-file-vaultik-destinationtxt).
+The `file://` backend never creates its destination directory, so a write to a
+missing one fails.
 
 The `{remote-key}` directory name is a one-way double SHA-256 hash of the human
 snapshot ID, so the human ID (hostname, snapshot name, timestamp) is never