fix(chain): handle long block height gap

[bowenwang1996]

Currently we drop blocks when they are more than 500 heights from our head. However, this means that if the network stalled for some time and recover, a valid block whose height is more than 500 ahead but whose prev block is our head may be dropped and will cause the node to stall.

Test plan

test_long_gap_between_blocks

[gitpod-io]

[codecov]

Codecov Report

Merging #3155 into master will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master    #3155   +/-   ##
=======================================
  Coverage   87.94%   87.94%           
=======================================
  Files         212      212           
  Lines       43004    43004           
=======================================
  Hits        37819    37819           
  Misses       5185     5185           

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update d2839ca...8ef1d5b. Read the comment docs.

[SkidanovAlex]

Can we test that if a block has height u64::max, we do not crash in any way before we reject it as having incorrect approvals?
Or keep the limit, but set it to something very large, like 4B?

[bowenwang1996]

@SkidanovAlex there is already a test for this https://github.com/nearprotocol/nearcore/blob/c9d37cfd7b07465f014aa64e198eb88bf6bd19be/chain/client/tests/process_blocks.rs#L976