BR: restore panic when backup has duplicate database and table name.

[3pointer]

Bug Report

Please answer these questions before submitting your issue. Thanks!

Normally we wouldn't have a duplicate table names in same cluster. but for system tables. we use temporary database __TiDB_BR_Temporary_mysql to store system tables. This may cause the problem.

1. Minimal reproduce step (Required)

1. run tiup playground
2. run create database __TiDB_BR_Temporary_mysql and create table tables_priv(id int) (anyone is as same as mysql database.)
3. run br backup full --pd xxx -s 'local://tmp/bk01'
4. run br restore full --pd xxx -s 'local://tmp/bk01'

2. What did you expect to see? (Required)

restore success

3. What did you see instead (Required)

panic due to we restore tables_priv twice.

panic: bug! table done before register!

goroutine 530 [running]:
go.uber.org/zap/zapcore.CheckWriteAction.OnWrite(0x2?, 0x2?, {0x0?, 0x0?, 0xc0006c33a0?})
	/go/pkg/mod/go.uber.org/zap@v1.23.0/zapcore/entry.go:198 +0x65
go.uber.org/zap/zapcore.(*CheckedEntry).Write(0xc001459930, {0xc0021ab000, 0x2, 0x2})
	/go/pkg/mod/go.uber.org/zap@v1.23.0/zapcore/entry.go:264 +0x3ec
go.uber.org/zap.(*Logger).Panic(0xc0021ab000?, {0x4cc400a?, 0x18c7ceb?}, {0xc0021ab000, 0x2, 0x2})
	/go/pkg/mod/go.uber.org/zap@v1.23.0/logger.go:251 +0x59
github.com/pingcap/log.Panic({0x4cc400a?, 0xe?}, {0xc0021ab000?, 0xc000bd59e0?, 0xc0ed4fb21799f4a0?})
	/go/pkg/mod/github.com/pingcap/log@v1.1.1-0.20221116035753-734d527bc87c/global.go:54 +0x8b
github.com/pingcap/tidb/br/pkg/restore.(*tikvSender).waitTablesDone(0xc0000fa9b0, {0xc0021d2600?, 0x9, 0xc001f7dba0?})
	/home/jenkins/agent/workspace/build-common/go/src/github.com/pingcap/br/br/pkg/restore/pipeline_items.go:351 +0x295
github.com/pingcap/tidb/br/pkg/restore.(*tikvSender).restoreWorker.func2()
	/home/jenkins/agent/workspace/build-common/go/src/github.com/pingcap/br/br/pkg/restore/pipeline_items.go:390 +0x452
golang.org/x/sync/errgroup.(*Group).Go.func1()
	/go/pkg/mod/golang.org/x/sync@v0.1.0/errgroup/errgroup.go:75 +0x64
created by golang.org/x/sync/errgroup.(*Group).Go
	/go/pkg/mod/golang.org/x/sync@v0.1.0/errgroup/errgroup.go:72 +0xa5

4. What is your TiDB version? (Required)

v6.5.0

[benmaoer]

Usually, when we find duplicated table/database names during restoring, there are two types of solutions:

• Option 1: prompt an error message to indicate the duplicated name, restore operation failed and let user to handle it.
• Option 2: prompt a warning, but continue to restore, automatically handle the duplicates ( rename to a new table / database ). Users are easy to decide which one to keep.