Add black_box to benchmarks

[sunsided]

This ensures the benchmarks do not accidentally optimize for the provided constants.

On my machine, this would change the reported benchmark results from

running 3 tests
test bench_all_price          ... bench:           2.44 ns/iter (+/- 0.79)
test bench_all_price_no_cache ... bench:          73.79 ns/iter (+/- 11.15)
test bench_call_price         ... bench:          43.34 ns/iter (+/- 14.57)

to

running 3 tests
test bench_all_price          ... bench:          61.22 ns/iter (+/- 30.21)
test bench_all_price_no_cache ... bench:         250.83 ns/iter (+/- 84.97)
test bench_call_price         ... bench:          40.98 ns/iter (+/- 21.33)

This should give a more accurate picture.

[coveralls]

Pull Request Test Coverage Report for Build 9243228171

Details

• 0 of 0 changed or added relevant lines in 0 files are covered.
• 1 unchanged line in 1 file lost coverage.
• Overall coverage decreased (-0.4%) to 80.163%

---

Files with Coverage Reduction	New Missed Lines	%
src/lib.rs	1	80.16%

Totals
Change from base Build 4354937295:	-0.4%
Covered Lines:	406
Relevant Lines:	445

---

💛 - Coveralls

[sunsided]

The benchmark test now obviously fails because the report baseline has changed.

[danielhstahl]

Didn't this get merged already? I'm surprised its not showing a merge conflict...

[sunsided]

@danielhstahl Must have been a glitch. I rebased it against master and now it's correctly an empty PR due to #28.

[danielhstahl]

Thanks @sunsided . Your changes have been deployed in 10.2: https://crates.io/crates/black_scholes