@peter/db split

[0g-peterzhb]

This PR does the following:

1. separate flow db from data db so the snapshot can contain just the flow db without the data db
2. data db contains actually data, miner_id, shard_config, finalized indicator
3. flow db contains merkle, sync event statistics
4. pad tx after the sync period finished to allow for faster sync up.

This change is 

[boqiu]

Reviewed 18 of 18 files at r2, all commit messages.
Reviewable status: complete! all files reviewed, all discussions resolved (waiting on @peilun-conflux)

[peilun-conflux]

Reviewed 1 of 5 files at r1, 18 of 18 files at r2, all commit messages.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @0g-peterzhb)

---

node/storage/src/log_store/log_manager.rs line 439 at r2 (raw file):

                        }
                        std::result::Result::Err(_) => {
                            debug!("Unable to get pad data, start_index={}", start_index);

This is possible when we reach the max tx seq, so maybe add a sleep here to avoid a spin loop?

[peilun-conflux]

Reviewed 1 of 1 files at r3, all commit messages.
Reviewable status: complete! all files reviewed, all discussions resolved (waiting on @0g-peterzhb)

[0g-peterzhb]

Reviewable status: complete! all files reviewed, all discussions resolved (waiting on @0g-peterzhb)

---

node/storage/src/log_store/log_manager.rs line 439 at r2 (raw file):

Previously, peilun-conflux wrote…

This is possible when we reach the max tx seq, so maybe add a sleep here to avoid a spin loop?

resolved.