temp file should be removed when sync interrupted.

[zhoucheng361]

What happened:

root@bench-01:~/juicefs# ./juicefs sync /tmp/jfs/bigfile /tmp/jfs/sync
2024/04/29 18:45:43.743250 juicefs[3859900] <INFO>: Prometheus metrics listening on 127.0.0.1:36403 [mount.go:135]
2024/04/29 18:45:43.745302 juicefs[3859900] <INFO>: Syncing from file:///tmp/jfs/bigfile to file:///tmp/jfs/sync [sync.go:1300]
Scanned objects: 0/1 [--------------------------------------------------------------]  0.0/s ETA: 0s
Skipped objects: 0                            0.0/s                                       ⠦
  Skipped bytes: 0.0 b   (0 Bytes)            0.0 b/s                                     ⠦
Pending objects: 0                            0.0/s                                       ⠦
 Copied objects: 0                            0.0/s                                       ⠦
   Copied bytes: 1.9 GiB (1994391552 Bytes)   337.2 MiB/s                                 ⠦
 Failed objects: 0                            0.0/s                                       ⠦
^C
root@bench-01:~/juicefs# ls /tmp/jfs/ -a
.  ..  abc  .accesslog  big_dir  bigdir2  bigfile  .config  fsrand  no_dismount.txt  .stats  .sync.tmp5862692998825721621  .trash  vdb.1_1.dir  vdb.1_2.dir
root@bench-01:~/juicefs# ls -lh /tmp/jfs/.sync.tmp5862692998825721621
-rw-r--r-- 1 root root 1.9G Apr 29 18:45 /tmp/jfs/.sync.tmp5862692998825721621

What you expected to happen:

How to reproduce it (as minimally and precisely as possible):

Anything else we need to know?

Environment:

• JuiceFS version (use juicefs --version) or Hadoop Java SDK version:
• Cloud provider or hardware configuration running JuiceFS:
• OS (e.g cat /etc/os-release):
• Kernel (e.g. uname -a):
• Object storage (cloud provider and region, or self maintained):
• Metadata engine info (version, cloud provider managed or self maintained):
• Network connectivity (JuiceFS to metadata engine, JuiceFS to object storage):
• Others: