br: should remove service-gc-safepoint when finished data import

## Bug Report

Please answer these questions before submitting your issue. Thanks!

### 1. Minimal reproduce step (Required)
1、 Use br import data;
[2022/03/17 13:27:59.512 +00:00] [INFO] [collector.go:68] ["Full Restore success summary"] [total-ranges=15867] [ranges-succeed=15867] [ranges-failed=0] [split-region=13m9.933041196s] [restore-ranges=15803] [total-take=1h5m30.884475391s] [restore-data-size(after-compressed)=523.9GB] [Size=523914531681] [BackupTS=431887417076875266] [total-kv=8960000000] [total-kv-size=1.138TB] [average-speed=289.5MB/s]
2、After import finished, hit following logs in tidb log:
[2022/03/17 13:31:56.356 +00:00] [INFO] [gc_worker.go:608] ["[gc worker] there's another service in the cluster requires an earlier safe point. gc will continue with the earlier one"] [uuid=5fe5efebd180015] [ourSafePoint=431888352209797120] [minSafePoint=431887417076875265]

### 2. What did you expect to see? (Required)
PD delete the service-gc-safepoint after TTL ( 5min in default )

### 3. What did you see instead (Required)
Br should delete it immediately after succeeded imported.

### 4. What is your TiDB version? (Required)

[2022/03/17 13:27:59.512 +00:00] [INFO] [collector.go:68] ["Full Restore success summary"] [total-ranges=15867] [ranges-succeed=15867] [ranges-failed=0] [split-region=13m9.933041196s] [restore-ranges=15803] [total-take=1h5m30.884475391s] [restore-data-size(after-compressed)=523.9GB] [Size=523914531681] [BackupTS=431887417076875266] [total-kv=8960000000] [total-kv-size=1.138TB] [average-speed=289.5MB/s]