API: add float128 and float32(64,128)_complex dt

[samnordmann]

What

Add support for four new datatypes: float128, float32_complex, float64_complex, float128_complex.

• float32_complex and float64_complex are supported on HOST and CUDA. float128 and float128_complex are supported only on HOST.
• The new datatypes are supported by all collectives. float128 supported with the reduction ops: min, max, sum, prod, avg. float32(64,128)_complex supported with the reduction ops: sum, prod, avg.
• Update MPI_test and gtest with the new datatypes. Remark: the gtests are run with float128 and float128 only on mc_reduce tests.
• new datatypes not supported by nccl nor sharp TLs

[manjugv]

@samnordmann It might be better to decouple the formatting changes from the functional changes.

[jladd-mlnx]

• Way too many white-space changes. I had to hide whitespaces in order to make sense of the PR.

• The file src/ucc/api/ucc.h is Whitespace-only changes.

• Should we include support for complex half precision? Does HPL AI use it?

typedef enum cudaDataType_t
{
        CUDA_R_16F= 2, // 16 bit real 
        CUDA_C_16F= 6, // 16 bit complex
        CUDA_R_32F= 0, // 32 bit real
        CUDA_C_32F= 4, // 32 bit complex
        CUDA_R_64F= 1, // 64 bit real
        CUDA_C_64F= 5, // 64 bit complex
        CUDA_R_8I= 3,  // 8 bit real as a signed integer 
        CUDA_C_8I= 7,  // 8 bit complex as a pair of signed integers
        CUDA_R_8U= 8,  // 8 bit real as a signed integer 
        CUDA_C_8U= 9   // 8 bit complex as a pair of signed integers
} cudaDataType;

[samnordmann]

@samnordmann It might be better to decouple the formatting changes from the functional changes.

• Way too many white-space changes. I had to hide whitespaces in order to make sense of the PR.
• The file src/ucc/api/ucc.h is Whitespace-only changes.

The formatting changes were due to a mistake with automatic formatting. It is fixed now.
@jladd-mlnx @manjugv

[Sergei-Lebedev]

not related to this PR, but there are SHARP_DTYPE_UINT8 and SHARP_DTYPE_UINT8 defined in sharp.h, should we update mapping? cc @bureddy

[vspetrov]

bot:retest

[vspetrov]

@samnordmann plz address final set of minor changes