[GPU] check the feature padding at onednn reorder

[sungeunk]

Description of the issue(symptom, root-cause, how it was resolved)

• onednn reorder can not handle the feature axis padding. 'is_supported_pad()' function does not check the feature axis padding.
• fallback to ocl if the reorder has the feature axis padding.

The code and line that caused this issue (if it is not changed directly)

• src/plugins/intel_gpu/src/graph/impls/onednn/reorder_onednn.cpp
• src/plugins/intel_gpu/src/graph/impls/onednn/utils.cpp

Reproduction step and snapshot (if applicable. Do not attach for customer model)

• reproducer is attached in the ticket

Problematic graph

Checklist

• Is it a proper fix? (not a workaround)
• Did you include test case for this fix, if necessary?
• Did you review existing test that can be extended to cover this scenario? Which test did you review?

Tickets:

• 172242

[Copilot]

Pull Request Overview

This PR fixes a bug in the OneDNN reorder implementation where feature axis padding was not being properly checked, causing unsupported operations to be attempted instead of falling back to OpenCL.

• Added feature axis padding validation to the is_supported_pad() function
• Modified the return condition to include the new feature padding check alongside existing spatial and batch padding checks

---

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

[geunhwan]

PR in the master branch has been approved.

[praasz]

Suggested change

	bool no_feature_padding = true;
	no_feature_padding &= (pad._lower_size[1] == 0);
	no_feature_padding &= (pad._upper_size[1] == 0);
	return no_feature_padding;
	return (pad._lower_size[1] == 0) && (pad._upper_size[1] == 0);

Can be refactored?

[p-durandin]

build_jenkins