* Integrate APU

* fix code style issue

Summary: Reformat fp16 gemv, add FP16_MATMUL_WEIGHT

Differential Revision: https://phabricator.d.xiaomi.net/D144134

1. Change DataFormat from enum to enum class.

1. Auto transformation between data format, data type and memory type.
2. Add device placement optimization strategy.

add extract pooling and kaldi batchnorm

add time index info in extract pooling

add benchmar& test for kaldi components

add batch support for kaldi

change ifdefined to delay

support kaldi tdnn-lstm

support logsoftmax

add ifdefined op

add dynamic lstm docs

1. Support int32 input data type.
2. Support GatherV2 op
3. Add transpose to ExpandDim op.

1. input_data_formats flag in yaml supports OIHW
2. MaceTensor support OIHW data format

1. Remove data_format flag for ambiguous
2. add has_data_format for distinguish models without data format.

support kaldi-onnx model convert
support validate output data from url

* Added floordiv() operator

* Restored original indentation

* Update eltwise.cl

Update eltwise.cl according to the recent fix.

1. Add registration of Op conditions.
2. Add data format for ops based on model's input data format.

add onnx_optimizer tool and update onnx info in docs

1. Support memory optimization at runtime
2. Support memory type transformation automatically at runtime.
3. Move gpu winograd convolution to Conv2D.
4. Support Data Type transformation automatically.
5. MaceTensor API support optional data type.
6. Add input_data_formats and output_data_formats at yaml
7. Related issue(#363)