perf(crypto): native bip38

[spkjp]

Proposed changes

The used bip38 library is abysmal slow.

For example take a look at the Delegate tests which invoke bip38 code:

Test Suites: 1 passed, 1 total
Tests:       14 passed, 14 total
Snapshots:   0 total
Time:        46.666s
Ran all test suites matching /__tests__\/models\/delegate.test.ts/i.

Even though they don't do anything demanding, they alone usually take 60-90 seconds to complete
when run by CI.

Looking over the bip38 code and open GH issues, it becomes apparent that it is mainly due to the used scryptsy library:

bitcoinjs/bip38#40
cryptocoinjs/scryptsy#8

So I decided to fork the bip38 code and made the following changes:

• use Node's own version of scrypt instead of scryptsy
• drop the used JavaScript ECC and use equivalent functions from the secp256k1 library

With my changes the same tests from above:

Test Suites: 1 passed, 1 total
Tests:       14 passed, 14 total
Snapshots:   0 total
Time:        2.983s, estimated 5s
Ran all test suites matching /__tests__\/models\/delegate.test.ts/i.

On my machine that is more than ~15x faster.

@arkecosystem/crypto now ships the modified bip38 version and core-forger and core-json-rpc were updated accordingly. It maybe makes sense to publish it as a standalone package so other people can use it as well, but since it uses some of our own crypto, I didn't bother. 🤷‍♂️

Types of changes

• Bugfix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Refactoring (improve a current implementation without adding a new feature or fixing a bug)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Build (changes that affect the build system)
• Docs (documentation only changes)
• Test (adding missing tests or fixing existing tests)
• Other... Please describe:

Checklist

• I have read the CONTRIBUTING documentation
• Lint and unit tests pass locally with my changes
• I have added tests that prove my fix is effective or that my feature works
• I have added necessary documentation (if appropriate)

[codecov-io]

Codecov Report

Merging #1941 into develop will increase coverage by 0.88%.
The diff coverage is 95.12%.

@@             Coverage Diff             @@
##           develop    #1941      +/-   ##
===========================================
+ Coverage    40.41%   41.29%   +0.88%     
===========================================
  Files          350      351       +1     
  Lines         7624     7742     +118     
  Branches      1143     1125      -18     
===========================================
+ Hits          3081     3197     +116     
- Misses        4528     4532       +4     
+ Partials        15       13       -2

Impacted Files	Coverage Δ
...son-rpc/src/server/methods/wallets/bip38/create.ts	0% <0%> (ø)	⬆️
...ages/core-json-rpc/src/server/utils/decrypt-wif.ts	0% <0%> (ø)	⬆️
packages/crypto/src/models/delegate.ts	78.26% <100%> (ø)	⬆️
packages/crypto/src/crypto/bip38.ts	96.66% <96.66%> (ø)
packages/core-logger-winston/src/formatter.ts	38.88% <0%> (ø)	⬆️
packages/core-test-utils/src/generators/wallets.ts	14.28% <0%> (ø)	⬆️

---

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 7d240d9...2ad8c65. Read the comment docs.