Add benchmark vs naked bunyan

[nason]

Resolves #58

npm run benchmark:

> we-js-logger@0.5.1 benchmark /Users/mnason/code/we-js-logger
> node test/benchmark

we-js-logger x 334,953 ops/sec ±2.09% (83 runs sampled)
we-js-logger child x 154,756 ops/sec ±2.25% (78 runs sampled)
bunyan x 322,868 ops/sec ±1.49% (83 runs sampled)
bunyan child x 229,026 ops/sec ±1.62% (87 runs sampled)
Fastest is we-js-logger

I was kind of surprised by this result (I expected bunyan to be faster). It might be because we're not configuring we-js-logger to do anything extra in this benchmark.

Still, this could be helpful with issues such as #54 in the future.

[ricardo-quinones]

👍

[coveralls]

Coverage remained the same at 77.174% when pulling bab02dc on benchmark into fc7b5bd on master.

[mattjstar]

Huh, that seems bizarre that we-js-logger would be faster, but it seems like the child is significantly slower or am I reading that incorrectly?

[nason]

Huh, that seems bizarre that we-js-logger would be faster, but it seems like the child is significantly slower or am I reading that incorrectly?

That's what I see too. Its not what I expected but at least we have some visibility into it now!

[mattjstar]

haha sounds good to me