[wip] ethdb: increase ideal batch size by 10x

[holiman]

Experiment with changing the ideal db batch size.

do not merge, this is just a PR for benchmarking.

[holiman]

Putting this on mon08/09 benchmarkers now, as a fast-sync.
https://geth-bench.ethdevops.io/d/Jpk-Be5Wk/dual-geth?orgId=1&var-exp=mon08&var-master=mon09&var-percentile=50&from=1571932500000&to=now&refresh=30s

[holiman]

Both were started (fast-sync) at 17:55,

both had downloaded all headers and receipts aroud 22:00

master finished fast-sync at 22:25 (4h 30m)
exp finished fast-sync at 23:25 (5h 30m)

exp:

Oct 24 23:28:03 mon08.ethdevops.io geth INFO [10-24|21:28:03.238] Imported new state entries count=13732 elapsed=121.149ms processed=383704532 pending=0 retry=0 duplicate=19358 unexpected=134657 

master:

Oct 24 22:23:17 mon09.ethdevops.io geth INFO [10-24|20:23:17.903] Imported new state entries count=17 elapsed=603.372µs processed=383555065 pending=0 retry=0 duplicate=6955 unexpected=44392 

the sync phase:

https://geth-bench.ethdevops.io/d/Jpk-Be5Wk/dual-geth?orgId=1&var-exp=mon08&var-master=mon09&var-percentile=50&from=1571932832513&to=1571958360740

In total, this PR didn't seem to make much difference either way