This should allow many algorithms to take advantage of more parallelization
technologies.

add hack to disable optimization of linear svms; improved precision of optimize_linear_svm; add the relevant test, which however requires some big database (so it's disabled by default)

Added new kernels to documentation

Histogram intersection kernel (accessible by using SVM::INTER in CV namespace as kernel_type)
Exponetial chi2 kernel (accessible by using SVM::CHI2 in CV namespace as kernel_type)

Formulars:
Exp-CHI2
k(x,y) = exp(-gamma * CHI2(x,y))
CHI2(x,y) = 1- 2* SUM_i[(xi-yi)²/(xi+yi)]

Intersec
k(x,y) = SUM_i[min(xi,yi)]

replaced alloca() (a.k.a. cvStackAlloc) with AutoBuffer or vector() everywhere. cvStackAlloc() is still defined, but we do not need alloca() anymore to compile and run OpenCV (fixes #889 and may be some others)