[clang-doc] Parallelize reducing phase

Reduce phase has been parallelized and a execution time was reduced by
60% with this.
The reading of bitcode (bitcode -> Info) was moved to this segment of
code parallelized so it now happens just before reducing.

Differential Revision: https://reviews.llvm.org/D65628

git-svn-id: https://llvm.org/svn/llvm-project/clang-tools-extra/trunk@368206 91177308-0d34-0410-b5e6-96231b3b80d8
1 file changed