A benchmark for our trigonometric functions and a few performance improvements

[pleroy]

• Support for angles in [-π/4, π/4].
• Support for uninitialized DoublePrecision objects.
• Improved computation of the table index.

Benchmark:

Run on (48 X 3793 MHz CPU s)
CPU Caches:
  L1 Data 32 KiB (x24)
  L1 Instruction 32 KiB (x24)
  L2 Unified 512 KiB (x24)
  L3 Unified 32768 KiB (x4)
------------------------------------------------------------------------------------------------------
Benchmark                                                            Time             CPU   Iterations
------------------------------------------------------------------------------------------------------
BM_EvaluateElementaryFunction<Metric::Latency, std::sin>          7.26 ns         7.25 ns    112000000 cycles: 27.5162
BM_EvaluateElementaryFunction<Metric::Throughput, std::sin>       2.12 ns         2.13 ns    344616000 cycles: 8.00978
BM_EvaluateElementaryFunction<Metric::Latency, cr_sin>            31.9 ns         31.5 ns     21334000 cycles: 120.947
BM_EvaluateElementaryFunction<Metric::Throughput, cr_sin>         20.8 ns         20.9 ns     34462000 cycles: 78.7098
BM_EvaluateElementaryFunction<Metric::Latency, Sin>               11.2 ns         11.2 ns     64000000 cycles: 42.4374
BM_EvaluateElementaryFunction<Metric::Throughput, Sin>            3.59 ns         3.61 ns    194783000 cycles: 13.5883
BM_EvaluateElementaryFunction<Metric::Latency, std::cos>          8.14 ns         8.02 ns     89600000 cycles: 30.8504
BM_EvaluateElementaryFunction<Metric::Throughput, std::cos>       2.12 ns         2.13 ns    344616000 cycles: 8.00876
BM_EvaluateElementaryFunction<Metric::Latency, cr_cos>            35.3 ns         35.3 ns     20364000 cycles: 133.698
BM_EvaluateElementaryFunction<Metric::Throughput, cr_cos>         21.7 ns         21.5 ns     32000000 cycles: 82.2311
BM_EvaluateElementaryFunction<Metric::Latency, Cos>               11.0 ns         11.0 ns     64000000 cycles: 41.5094
BM_EvaluateElementaryFunction<Metric::Throughput, Cos>            3.28 ns         3.30 ns    213334000 cycles: 12.4005

#1760.