GitLab

Instead of replacing all #ifdef USE_CUDA, just add USE_CUDA define to ROCm build.

Previously #6086 added ROCm support but after numerous rebases it lost
critical changes. This PR restores the ROCm build.

There are many source file changes but most were automated using the
following:

```bash
for f in `grep -rl '#ifdef USE_CUDA'`
do
    sed -i 's@#ifdef USE_CUDA@#if defined(USE_CUDA) || defined(USE_ROCM)@g' $f
done

for f in `grep -rl '#endif  // USE_CUDA'`
do
    sed -i 's@#endif  // USE_CUDA@#endif  // USE_CUDA || USE_ROCM@g' $f
done
```

* [ROCm] add support for ROCm/HIP

- CMakeLists.txt ROCm updates, also replace glob with explicit file list
- initial warpSize interop changes
- helpers/hipify.sh script added
- .gitignore to ignore generated hip source files

* more rocm updates

- disable compiler warnings
- move PercentileDevice __device__ template function into header
- bug fixes for __host__ __define__ and __HIP__ preprocessor symbols

* more bug fixes

* warp 32 vs 64 updates

* lint fixes

* missing device_index variable

* accidental inclusion of hip headers

* copyright notice compliance

* Update CMakeLists.txt
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* fix lint issue

* clean up

* Update CMakeLists.txt
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* Update CMakeLists.txt
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* clean up CMakeLists.txt

use WARPSIZE

* use WARPSIZE

* fix share buffer size

---------
Co-authored-by: shiyu1994 <shiyu_k1994@qq.com>
Co-authored-by: James Lamb <jaylamb20@gmail.com>
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>
Co-authored-by: Yu Shi <yushi2@microsoft.com>

* add cuda quantile regression objective

* remove white space

* resolve merge conflicts

* remove useless changes

* remove useless changes

* enable cuda quantile regression objective

* add a test case for quantile regression objective

* remove useless changes

* remove useless changes

* reduce DP_SHARED_HIST_SIZE to 5176 for CUDA 10

---------
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* [ci] speed up if-else, swig, and lint conda setup

* add 'source activate'

* python constraint

* start removing cuda v1

* comment out CI

* remove more references

* revert some unnecessaary changes

* revert a few more mistakes

* revert another change that ignored params

* sigh

* remove CUDATreeLearner

* fix tests, docs

* fix quoting in setup.py

* restore all CI

* Apply suggestions from code review
Co-authored-by: shiyu1994 <shiyu_k1994@qq.com>

* Apply suggestions from code review

* completely remove cuda_exp, update docs

---------
Co-authored-by: shiyu1994 <shiyu_k1994@qq.com>

* add rmse metric for new cuda version

* add Init for CUDAMetricInterface

* fix lint errors

* fix rmse and add l2 metric for new cuda version

* use CUDAL2Metric

* explicit template instantiation

* write result only with the first thread

* pre allocate buffer for output converting

* fix l2 regression with cuda metric evaluation

* weighting loss in cuda metric evaluation

* mark CUDATree::AsConstantTree as override

* add rmse metric for new cuda version

* add Init for CUDAMetricInterface

* fix lint errors

[CUDA] Add Poisson regression objective for cuda_exp and refactor objective functions for cuda_exp (#5486)

* add poisson regression objective for cuda_exp

* enable Poisson regression for cuda_exp

* refactor cuda objective functions

* remove useless changes

* fix linter errors

* remove redundant buffer in cuda poisson regression objective

* fix log of cuda_exp binary objective

* fix threshold of poisson objective result

* remove useless changes

* fix compilation errors

* add cuda quantile regression objective

* remove cuda quantile regression objective
Co-authored-by: James Lamb <jaylamb20@gmail.com>

* add multiclass objective for cuda_exp

* remove debug code

* add includes requested by lint checks

* fix compilation failure for cuda with cuda-9.0

* clean code

* change percentile in CUDARegressionL1loss::BoostFromScore to 0.5

* add lambdarank for cuda_exp

* support unlimited number of ranks in labels

* fix lint errors

* remove warning for lambdarank with cuda_exp

* Update src/objective/cuda/cuda_rank_objective.hpp
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* Update src/objective/cuda/cuda_rank_objective.hpp
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>

* Update cuda_column_data.hpp

* Update cuda_metadata.hpp

* Update cuda_objective_function.hpp

* Update cuda_row_data.hpp

* Update cuda_regression_objective.hpp

* add huber regression for cuda_exp

* renew tree output on GPU

add test cases for regression objectives

* remove useless changes

* add white space

* fix test_regression

* add (l1) regression objective for cuda_exp

* remove RenewTreeOutputCUDA from CUDARegressionL2loss

* remove mutable and use CUDAVector

* remove white spaces

* remove TODO and document in (#5459)

* add (l2) regression objective for cuda_exp

* fix lint errors

* correct time tag

* add binary objective for cuda_exp

* include <string> and <vector>

* exchange include ordering

* fix length of score to copy in evaluation

* fix EvalOneMetric

* fix cuda binary objective and prediction when boosting on gpu

* Add white space

* fix BoostFromScore for CUDABinaryLogloss

update log in test_register_logger

* include <algorithm>

* simplify shared memory buffer