fix: eth fetcher re-queue issue

[galaio]

Description

Some node report sync sucks, node logs as below:

t=2023-02-01T07:44:23+0000 lvl=info msg="Imported new chain segment" blocks=1 txs=138 mgas=18.556 elapsed=2.235s mgasps=8.302 number=25,284,893 hash=0x3b08654acd530ea6d793610218f81d9614d3b31b8eeffe7b372fc809ce15fc09 dirty="226.27 MiB" 
t=2023-02-01T07:44:23+0000 lvl=info msg="Chain reorg detected" number=25,284,892 hash=0x331cb0211dab8c564d87583bcf540ba4030b6b69c4e7d6887252b319ed940914 drop=1 dropfrom=0x3b08654acd530ea6d793610218f81d9614d3b31b8eeffe7b372fc809ce15fc09 add=1 addfrom=0xc91bb810c5aca2c562d36335835673f119b2cdf7404ad18e1b1ff223b9ceada2 
t=2023-02-01T07:44:23+0000 lvl=info msg="Imported new chain segment" blocks=1 txs=123 mgas=20.816 elapsed=226.522ms mgasps=91.893 number=25,284,893 hash=0xc91bb810c5aca2c562d36335835673f119b2cdf7404ad18e1b1ff223b9ceada2 dirty="226.34 MiB" 
t=2023-02-01T07:44:23+0000 lvl=info msg="Re-queue blocks" number=25,284,894 hash=0x3247b6f9fdacd178233a3f77dd7402789983590920974110e70b1dfb12e9b16a 
t=2023-02-01T07:44:34+0000 lvl=info msg="Deep froze chain segment" blocks=16 elapsed=42.439ms number=25,194,893 hash=0xeee1826735b032aea70f6387074c59aeae42c9fec404738fae02666747933371 
t=2023-02-01T07:45:19+0000 lvl=info msg="Downloader queue stats" receiptTasks=0 blockTasks=0 itemSize="26.32 KiB" throttle=8192

In the eth/fetcher main loop, it pops block order by block number, when a poped block height = chainHeight+1, fetcher will insert it into the chain.

Let's think about this situation:

If two continuous blocks arrived, the chain height = the first block height, the first block causes a chain reorg, next block could insert concurrently, but got a Re-queue block alert, and try re-queue. Because the queued state had recorded it before, the re-queue action failed.

Rationale

The PR opt re-queue logic, force block enqueue and wait for the next pop.

Changes

Notable changes:

• fix the re-queue failed bug;