make format

Strided transpose element restriction does not require atomic adds

Minor function signature fix caught with ASan

memcheck - match nan check for QFunction field size

cpu - fix memcheck detection of uninitalized outputs

swarm - fix op swarm input access

Replace calls to deprecated `local_accessor::get_pointer()`

Merge pull request #1300 from CEED/sjg/hcurl-hdiv-basis-cuda-hip

H(div) and H(curl) basis support for `cuda-ref` and `hip-ref` backends

OperatorAtPoints (#1390)

* op - add AtPoints interface

* op - reorder AtPoints QFunction for consistency

* op - add CEED_EVAL_WEIGHT for operator AtPoints

* doc - add CeedOperatorCreateAtPo

OperatorAtPoints (#1390)

* op - add AtPoints interface

* op - reorder AtPoints QFunction for consistency

* op - add CEED_EVAL_WEIGHT for operator AtPoints

* doc - add CeedOperatorCreateAtPoints to release notes

show more ...

rstr - reorder AtPoints ordering for consistency

H(div) and H(curl) basis support for cuda-ref and hip-ref backends

Remove legacy CeedChk and CeedChkBackend (should use CeedCall, CeedCallBackend instead)

Add updated MI100 tuning data; make style updates for tuning data

Add updated A100 tuning data

Add updated MI250X tuning data

Add RTC tuning data for V100 and H100

Increment NB by 2 or 4 to speed up tuning

Add 2*p quadrature rules to tuning data (previously was just 2*(p-1)

Add documentation for MAGMA autotuning

New RTC tuning data format

Fixes and improvements to autotuning code

Add Python program to automatically generate tuning data for MAGMA RTC kernels

Add simple test program to generate benchmark data for CeedBasis operations

Whitespace formatting consistency with libCEED style

Remove need to zero out V vector before applying basis transpose for magma backends