Created Abstract class DFG2LLVM. The two passes DFG2LLVM_NVPTX and DFG2LLVM_X86 inherit from this class. Common features/functions have been put in the abstract class DFG2LLVM

- Bug fix in sgemm. opencl_nvidia
- Changed options.py to not clean (to keep .ll vectorized file in visc_tc_vec)
- Updated parboil print script

Do not allow argument type mungling for __visc__ calls. Changes to lib/Transforms/InstCombine/InstCombineCalls.cpp

Revert "Added a patch approved in future llvm branches which stop argument promotion for variagic functions. This was required to stop auto promotion of arguments in __visc__ calls"

This reverts commit f960f4d0.

This reverts commit 8938b17c.

Added a patch approved in future llvm branches which stop argument promotion for variagic functions. This was required to stop auto promotion of arguments in __visc__ calls

code enquires about its ancestor. Enables to run visc parboil benchmarks on x86
alone

(2) Modified unittests to have llvm.visc.init and cleanup intrinsics. Now they
all pass

(3) Modified visc.mk to easily compile visc version for just x86. Pass
TARGET=x86 as option to make command

(2) Added a compiled ptx assembly file for sgemm. Working for sgemm