tools: fix lagging cinfo.AppParams data structure in PingPong

[algoidurovic]

Summary

This PR intends to fix the following issues in PingPong's pps.prepareApps():

• pps.cinfo.AppParams is used in multiples places (ex. terminating from the app creation loop and appID access in the opt in loop) where its contents lag behind what the code expects. The solution is to either block until there are no longer apps pending creation and then update AppParams or remove the reference to AppParams entirely.
• The app creation loop only allows for one app per account, meaning PingPong won't support more apps than accounts.

Test Plan

Tested on local Algorand network.

[codecov]

Codecov Report

Merging #4613 (a819bb3) into master (5be155b) will increase coverage by 0.04%.
The diff coverage is 0.00%.

@@            Coverage Diff             @@
##           master    #4613      +/-   ##
==========================================
+ Coverage   54.20%   54.25%   +0.04%     
==========================================
  Files         402      402              
  Lines       51830    51849      +19     
==========================================
+ Hits        28096    28130      +34     
+ Misses      21370    21358      -12     
+ Partials     2364     2361       -3     

Impacted Files	Coverage Δ
shared/pingpong/accounts.go	3.27% <0.00%> (-0.14%)	⬇️
crypto/merkletrie/trie.go	66.42% <0.00%> (-2.19%)	⬇️
crypto/merkletrie/node.go	91.62% <0.00%> (-1.87%)	⬇️
network/wsNetwork.go	64.63% <0.00%> (-0.20%)	⬇️
ledger/catchpointtracker.go	62.89% <0.00%> (ø)
ledger/accountdb.go	73.03% <0.00%> (+0.15%)	⬆️
catchup/service.go	69.38% <0.00%> (+0.49%)	⬆️
ledger/acctonline.go	78.64% <0.00%> (+0.52%)	⬆️
ledger/acctupdates.go	70.19% <0.00%> (+0.59%)	⬆️
agreement/proposalManager.go	98.03% <0.00%> (+1.96%)	⬆️
... and 4 more

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[brianolson]

This loses the possibility that accounts with apps already exist. One of the big changes in my recent pingpong hacking was that we can now have stable accounts and apps and assets that exist over multiple invocations of pingpong. Maybe existingAppCnt := len(pps.cinfo.AppParams) and then loop while newAppCnt + existingAppCnt < pps.cfg.NumApp

[algoidurovic]

Ah got it. That explains a few things I was confused about, like why some failing pingpong invocations worked on subsequent runs. Your suggestion makes sense.