Fixed an encoder debug/relese mismatch in x86_64-win64-vs8

Fixed an encoder debug/relese mismatch in x86_64-win64-vs8

Visual c++ compiler uses xmm registers for floating point
operations for 64 bit architecture, therefore its calling
convention requires the preservation of xmm6-xmm15 in any
function that have used these registers. However, the sse2
functions, that were originally written for 32 bit windows,
may have used xmm6 and xmm7 without preserving the content.
In this particular case, the compiler used xmm6 to save
the variable "two_pass_min_rate", the value of the variable
is mucked up by our sse2 optimized loop filter functions,
hence the results of release/debug mismatching.