Borrow the library to index it, and give it back

The passes that need every photograph's bytes — thumbnails, face
indexing — now borrow each one and release it at the end. On a
placeholder library that is the difference between peak disk being the
working set and being the whole library.

Including on cancellation, which was nearly missed: the face sweep
returns mid-loop when the user presses Stop, and without releasing there
the disk is spent and nothing is delivered for it.

`materialise` now answers whether *it* fetched the content. The pool used
to work that out by listing a file's parent directory — one listing per
file across a library — when the backend already had to `stat` it to
decide whether to ask. One syscall instead of a directory walk, and it
removes the bug class the tests found earlier: a file at the library root
has no `parent()`, so every one of them read as already-downloaded.

**Pinning is the retention control**, and it drives the model the catalog
already had rather than a second one. `tier_desired` is what the user
asked to keep hydrated, `pending_pins` is the resumable work list, and a
pinned collection is never dehydrated for the same reason it was never
evicted. It was in fact *broken* here before: `get` on a stub failed, and
the pin worker logged "one unreadable file must not abandon the whole
pin" and silently did nothing.

Pinned originals on such a library are recorded with `path = NULL`
(`Cache::record_in_place`) rather than copied under `originals/`. Two
reasons, and the second is the important one. A copy would hold every
pinned photograph twice, with the budget able to evict the half that was
not costing the disk. And `release` deletes the file a row names — so a
row that names none cannot delete anything, which puts the one
catastrophic operation out of reach by construction rather than by
remembering not to call it. Deleting a materialised file inside a synced
tree removes the photograph from the server and every other device.

Handing disk back is `spawn_dehydrate`, which asks the client.

Two gaps written down rather than papered over (docs/storage.md §7): a
hydrating pass cannot yet quote its cost, because a stub reports no size;
and the two sweeps hold separate pools, so a library indexed for both
fetches twice.

core/dr-sync-folder/src/lib.rs

@@ -773,12 +773,12 @@ impl RemoteBackend for FolderBackend {
     /// Suffix-mode VFS *renames* on hydration, so completion is the
     /// materialised path appearing — not the stub changing size. Polling the
     /// original would wait forever.
-    async fn materialise(&self, id: &RemoteId) -> Result<(), RemoteError> {
+    async fn materialise(&self, id: &RemoteId) -> Result<bool, RemoteError> {
         let (local, materialised) = self.locate_id(id)?;
         if materialised {
-            // Already here. Not an error, and not a reason to ask again: the
-            // borrow pool relies on this being idempotent.
-            return Ok(());
+            // Already here. Not an error, and not a reason to ask again — and
+            // `false` is what tells a borrower to leave it alone afterwards.
+            return Ok(false);
         }
         let vfs = self.vfs.clone();
         let target = self.resolve_id(id)?;
@@ -793,7 +793,7 @@ impl RemoteBackend for FolderBackend {
             let deadline = std::time::Instant::now() + MATERIALISE_TIMEOUT;
             while std::time::Instant::now() < deadline {
                 if target.is_file() {
-                    return Ok(());
+                    return Ok(true);
                 }
                 std::thread::sleep(POLL);
             }