[PATCH v1 proxmox-backup 1/2] datastore: improve error if prune can't lock a snapshot for deletion

[Robert Obkircher <r.obkircher@proxmox.com>]

Include the locking failure in the message (without causes to avoid
duplicate paths) and switch to a more readable representaiton of the
BackupDir.

Changes the error message from:

  2026-04-29T12:16:17+02:00: failed to remove dir "host/19/2026-04-29T10:16:13Z": while destroying snapshot 'BackupDir { store: "data3", ns: BackupNamespace { inner: [], len: 0 }, dir: BackupDir { group: BackupGroup { ty: Host, id: "19" }, time: 1777457773 }, backup_time_string: "2026-04-29T10:16:13Z" }'

to

  2026-04-29T16:38:24+02:00: failed to remove dir "host/28/2026-04-29T14:38:19Z": while destroying snapshot "/srv/aaaa/host/28/2026-04-29T14:38:19Z": unable to acquire snapshot lock "/run/proxmox-backup/locks/aaaa/host-28-2026\\x2d04\\x2d29T14\\x3a38\\x3a19Z"

This error is reproducible with PBS in a single-core VM by running
"Prune All" while starting many concurrent backups.

Signed-off-by: Robert Obkircher <r.obkircher@proxmox.com>
---
 pbs-datastore/src/backup_info.rs | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/pbs-datastore/src/backup_info.rs b/pbs-datastore/src/backup_info.rs
index d3871933b..1349650ac 100644
--- a/pbs-datastore/src/backup_info.rs
+++ b/pbs-datastore/src/backup_info.rs
@@ -886,9 +886,9 @@ impl BackupDir {
     pub(crate) fn destroy(&self, force: bool, backend: &DatastoreBackend) -> Result<(), Error> {
         let (_guard, _manifest_guard);
         if !force {
-            _guard = self
-                .lock()
-                .with_context(|| format!("while destroying snapshot '{self:?}'"))?;
+            _guard = self.lock().map_err(|err| {
+                format_err!("while destroying snapshot {:?}: {err}", self.full_path())
+            })?;
             _manifest_guard = self.lock_manifest()?;
         }
 
-- 
2.47.3