Cache creators in cow base

[tolikzinovyev]

Summary

The evaluator shouldn't ask the ledger who is the creator of an asset multiple times. This PR adds a cache for ledger's responses in cow base.

Test Plan

Added a test. Is there a better way to write a mock test?

[codecov-commenter]

Codecov Report

Merging #2939 (4420f23) into master (01d853f) will increase coverage by 0.03%.
The diff coverage is 63.63%.

@@            Coverage Diff             @@
##           master    #2939      +/-   ##
==========================================
+ Coverage   47.27%   47.31%   +0.03%     
==========================================
  Files         356      356              
  Lines       57239    57249      +10     
==========================================
+ Hits        27062    27089      +27     
+ Misses      27108    27099       -9     
+ Partials     3069     3061       -8     

Impacted Files	Coverage Δ
ledger/appcow.go	81.95% <0.00%> (-0.25%)	⬇️
ledger/evalIndexer.go	38.55% <ø> (ø)
ledger/eval.go	78.03% <70.00%> (-0.15%)	⬇️
ledger/roundlru.go	90.56% <0.00%> (-5.67%)	⬇️
network/wsPeer.go	72.14% <0.00%> (-2.23%)	⬇️
network/wsNetwork.go	60.90% <0.00%> (ø)
ledger/acctupdates.go	63.01% <0.00%> (+0.51%)	⬆️
network/requestTracker.go	71.12% <0.00%> (+0.86%)	⬆️
catchup/peerSelector.go	100.00% <0.00%> (+1.04%)	⬆️
ledger/msgp_gen.go	35.32% <0.00%> (+1.12%)	⬆️
... and 4 more

---

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 01d853f...4420f23. Read the comment docs.

[tsachiherman]

looks good to me.

[algorandskiy]

Looks good, only concerning on FoundAddress type name

[algorandskiy]

why is it exported? The struct name sound a bit weird. Maybe cachedCreator or something?

[tolikzinovyev]

FoundAddress is also used in evalIndexer.go where this type needs to be exposed to Indexer. I think this name is ok for caching as well. FoundAddress encodes an address that is found (or not found). That is exactly what we want to cache.

[tsachiherman]

When do we need to cache non-existent addresses ? It looks like the Exists it's a mutually exclusive with the Address.

[tolikzinovyev]

Maybe caching responses for non-existent assets is silly, but it lets indexer preload such responses in a batch.

[tsachiherman]

I think we can do better here on both the struct naming as well on the cached content, but I can't see a (good) reason to block this one right away. We could iterate on the above in the future.