Try to make onnxruntime pass on OpenVINO

[sgbihu]

Description

• QuantizeLiner result different with 1 bias: Part of QuantizeLinear results have a bias of 1 microsoft/onnxruntime#25045, MS has 1 updates. May need update the test to accept the threshold. Need to change the test case.
• Float input/output: Test has construct float input and output nodes microsoft/onnxruntime#25044
• Resize uint8 accuracy issue: The resize operator's behavior may use Round instead of Floor microsoft/onnxruntime#25417, change test threshold
• Slice INT_MAX to INT_MIN: https://jira.devtools.intel.com/browse/CVS-169019, OV PR30906, DOC: For slicing to the end of a dimension with unknown size, it is recommended to pass in INT_MAX when slicing forward and ‘INT_MIN’ when slicing backward.

Motivation and Context

[Copilot]

Pull Request Overview

This PR addresses multiple OpenVINO execution provider compatibility issues to make onnxruntime tests pass. The changes primarily involve adjusting test tolerances and adding conditional skips for known OpenVINO limitations.

• Updates test error thresholds for quantization and resize operations to accommodate OpenVINO's precision differences
• Adds conditional test skips for OpenVINO when handling edge cases like empty tensors and negative slicing
• Enhances tensor checking infrastructure to support absolute error validation for integer data types

Reviewed Changes

Copilot reviewed 8 out of 8 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
slice_op.test.cc	Adds OpenVINO skip for negative step slicing with INT_MAX ends
resize_op_test.cc	Increases absolute error tolerance for uint8 resize operations
quantize_linear_test.cc	Sets absolute error tolerance for various quantization data types
concat_op_test.cc	Excludes OpenVINO from tests with zero-size inputs
cast_op_test.cc	Adds OpenVINO skip for empty tensor casting
loop_test.cc	Excludes OpenVINO from floating node tests
checkers.cc	Enhances Int4x2, UInt4x2, uint8_t, and uint16_t tensor checkers with absolute error support
quantize_ops_test.cc	Sets absolute error tolerance for contrib quantization operations

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}