This reverts commit 225da022.

merge with itseez/opencv/master

Jpeg2k color to greyscale conversion on non-Windows is done post decoding because system libjasper segfaults when decoding color images as greyscale.

On Windows, the tmpnam function returns a temp filename in the current directory, which has a prepended backslash '\\'.
This subsequently fails the open function.

This patch creates a proper temp filename in the temp folder and makes unlike work by opening the file as short-lived.

small formatting fixes

changed tabs in last commit to spaces

do not use this variable if it is not an tilted feature

This allows to start run.py with --move_tests parameter to place tests

to another location and keep them runnable from there

--help message and starting run.py w/o parameters generate different messages

E. g. w/ and w/o build_path in the end

build_path is required, removing square brackets

 - Made use imread() in OcvImageProcessing sample.
 - Added test/sample logic to verify file creation via imgcodecs.
Signed-off-by: Maxim Kostin <v-maxkos@microsoft.com>

    * correct the part about fixed aspect
    * improve formatting
    * add snippet markers to source
    * replace inline code by @snippet
  - do not run calibration twice when using a imageList
  - make output consistent in itself (CamelCase vs no_camel_case) as well as with old camera calibrate sample
  - respect write extrinsic/ points flags
  - set the aspectRatio value when specified
  - fix writing of calibration settings. also update grammar.
  - fix intendation and remove some size_t -> int casts by using size_t

 - Resolved GET_ENV and input type incompatibility
 - Made libjpeg compile for WINRT. So does imgcodecs module.
 - Updated .gitignore (log, tlog)

How this worked before, I do not know.

The check for BUILD_SHARED_LIBS had its logic backwards. If this
variable is not defined we must assume a build of static libs.

Make detection of x64 using the gcc's target triplet identical
to the one used in cmake/OpenCVDetectCXXCompiler.cmake.
Otherwise, MinGW-w64 setups will always be treated as x64 since
they contain "w64" as vendor key.

Fix typo that would always lead to detection of x86 for MinGW
builds in the OpenCVConfig.cmake file.

MinGW-w64 always uses "w64" as vendor key which the previously
used check for "64" anywhere in the target triplet matched. This
would lead to MinGW-w64 setups always being treated as x64.

Turns out we do not even need this additional check since the
architecture has been correctly determined earlier in this file.
No need to do it again.