[RFC] Use EPS instead of 0.1 as lower limit in non-zero positive constants present in C benchmark files

[manit2004]

In some statistical distributions 0.1 has been used as a lower bound for non zero positive numbers instead of EPS defined in stdlib/constants/float64/eps.h. I think stats packages should use EPS instead of hard coded constant like 0.1. Here are the list of benchmark files affected by it.

Weibull:

lib/node_modules/@stdlib/stats/base/dists/weibull/stdev/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/weibull/quantile/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/weibull/entropy/benchmark/c/benchmark.c

Student's t:

lib/node_modules/@stdlib/stats/base/dists/t/pdf/benchmark/c/benchmark.c

Poisson:

lib/node_modules/@stdlib/stats/base/dists/poisson/mgf/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/poisson/cdf/benchmark/c/benchmark.c

Planck:

lib/node_modules/@stdlib/stats/base/dists/planck/variance/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/planck/stdev/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/planck/skewness/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/planck/quantile/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/planck/pmf/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/planck/median/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/planck/mean/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/planck/mgf/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/planck/logcdf/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/planck/cdf/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/planck/kurtosis/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/planck/logpmf/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/planck/entropy/benchmark/c/benchmark.c

Pareto Type 1:

lib/node_modules/@stdlib/stats/base/dists/pareto-type1/variance/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/pareto-type1/pdf/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/pareto-type1/mode/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/pareto-type1/logcdf/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/pareto-type1/entropy/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/pareto-type1/cdf/benchmark/c/benchmark.c

Normal:

lib/node_modules/@stdlib/stats/base/dists/normal/stdev/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/normal/skewness/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/normal/mode/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/normal/median/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/normal/mean/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/normal/logcdf/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/normal/kurtosis/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/normal/entropy/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/normal/cdf/benchmark/c/benchmark.c

Negative Binomial:

lib/node_modules/@stdlib/stats/base/dists/negative-binomial/stdev/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/negative-binomial/variance/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/negative-binomial/skewness/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/negative-binomial/mgf/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/negative-binomial/mode/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/negative-binomial/mean/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/negative-binomial/kurtosis/benchmark/c/benchmark.c

Lognormal:

lib/node_modules/@stdlib/stats/base/dists/lognormal/pdf/benchmark/c/benchmark.c

Levy:

lib/node_modules/@stdlib/stats/base/dists/levy/cdf/benchmark/c/benchmark.c

Kumaraswamy:

lib/node_modules/@stdlib/stats/base/dists/kumaraswamy/skewness/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/kumaraswamy/logcdf/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/kumaraswamy/logpdf/benchmark/c/benchmark.c

Inverse Gamma:

lib/node_modules/@stdlib/stats/base/dists/invgamma/cdf/benchmark/c/benchmark.c

Chi:

lib/node_modules/@stdlib/stats/base/dists/chi/logpdf/benchmark/c/benchmark.c
lib/node_modules/@stdlib/stats/base/dists/chi/pdf/benchmark/c/benchmark.c

Chi-square:

lib/node_modules/@stdlib/stats/base/dists/chisquare/mgf/benchmark/c/benchmark.c

[kgryte]

The use of 0.1 is a valid value and we don't need to bring in the eps dependency for these files.