memdb: use atomic u64 addr to reduce allocation

[you06]

After #1389, we cache the last traversed node in memdb, but it's an atomic pointer, which need a heap allocation every time we update it, so the benchmark has little regression.

go test ./internal/unionstore -v -benchmem -bench=BenchmarkMemDbBufferSequential -run=^a -benchtime=1x

before  #1389
BenchmarkMemDbBufferSequential-8               1          42312917 ns/op        12576704 B/op         31 allocs/op

after #1389
BenchmarkMemDbBufferSequential-8               1          46407916 ns/op        15776704 B/op     200031 allocs/op

this pr
BenchmarkMemDbBufferSequential-8               1          44018167 ns/op        12576704 B/op         31 allocs/op

This PR also fixes some bugs of memdb:

• The refactor leave a mistake of the snapshot reverse iterator.
• The cached addr should be reset when memdb is reset.

[you06]

@cfzjywxk PTAL

[ti-chi-bot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: cfzjywxk, ekexium

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [cfzjywxk,ekexium]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[ti-chi-bot]

[LGTM Timeline notifier]

Timeline:

• 2024-08-27 07:27:11.651498102 +0000 UTC m=+855226.785948223: ☑️ agreed by ekexium.
• 2024-08-28 03:59:13.537259489 +0000 UTC m=+929148.671709610: ☑️ agreed by cfzjywxk.