Added CLEAN_JS and MINI_JS benches

[neeldug]

These benches are arbitrary code which is subject to change,
functionally the programs are identical.

This Pull Request fixes/closes #564 .

It changes the following:

• Exec and Full benchmarks to run with minified and clean code for comparison.

Specifically went for slightly longer function for visible change when minifying code with whitespace and character savings.

[codecov]

Codecov Report

Merging #581 into master will increase coverage by 1.30%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master     #581      +/-   ##
==========================================
+ Coverage   71.25%   72.56%   +1.30%     
==========================================
  Files         176      179       +3     
  Lines       11557    13259    +1702     
==========================================
+ Hits         8235     9621    +1386     
- Misses       3322     3638     +316     

Impacted Files	Coverage Δ
boa/src/builtins/object/internal_state.rs	0.00% <0.00%> (-41.67%)	⬇️
boa/src/builtins/value/rcstring.rs	69.23% <0.00%> (-7.70%)	⬇️
boa/src/builtins/number/mod.rs	62.28% <0.00%> (-6.16%)	⬇️
boa/src/builtins/error/mod.rs	32.00% <0.00%> (-4.37%)	⬇️
boa/src/builtins/object/mod.rs	38.11% <0.00%> (-4.16%)	⬇️
boa/src/builtins/value/conversions.rs	49.27% <0.00%> (-4.06%)	⬇️
boa/src/builtins/value/mod.rs	65.51% <0.00%> (-3.64%)	⬇️
boa/src/builtins/json/mod.rs	79.48% <0.00%> (-1.85%)	⬇️
boa/src/builtins/array/mod.rs	79.26% <0.00%> (-1.44%)	⬇️
boa/src/builtins/mod.rs	22.58% <0.00%> (-0.76%)	⬇️
... and 29 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 13f7374...9f623b1. Read the comment docs.

[HalidOdat]

Should this really go in exec? this to me seems like we are benchmarking the lexer/parser and not execution.

What do you think? @Razican @jasonwilliams

[Razican]

Should this really go in exec? this to me seems like we are benchmarking the lexer/parser and not execution.

If I'm not mistaken, the part inside of the bench() iterator is the execution (or the full thing), so they are in the correct place, but it would be very nice to add parsing benchmarks too.

[neeldug]

Should this really go in exec? this to me seems like we are benchmarking the lexer/parser and not execution.

If I'm not mistaken, the part inside of the bench() iterator is the execution (or the full thing), so they are in the correct place, but it would be very nice to add parsing benchmarks too.

I can implement this in the same PR, if this is wanted?

[Razican]

I can implement this in the same PR, if this is wanted?

Yes, please, if you can, that would be more complete :)

[Razican]

I think this is blocked until #640 lands, since it will need a refactor.

[Lan2u]

#640 is now merged so this is unblocked :)

[Razican]

This is good from my side, but let's wait for #646 to land in order to not fail the tests.