Fix issue where calculation of valuetypeshapecharacteristic on Arm64 does not match native runtime

[davidwrighton]

• We didn't check to make sure that the type of the Vector was a primitive numeric as the runtime does
• Re-enable test

Fixes#60036

[trylek]

LGTM, thanks for fixing this!

[AntonLapounov]

Vector<bool> is not supported and any operation on it will throw. Why do we even allow this type? @tannergooding

runtime/src/libraries/System.Private.CoreLib/src/System/Numerics/Vector_1.cs

Lines 155 to 170 in e2511ae

	publicstaticboolIsSupported
	{
	[MethodImpl(MethodImplOptions.AggressiveInlining)]
	get=>(typeof(T)==typeof(byte))||
	(typeof(T)==typeof(double))||
	(typeof(T)==typeof(short))||
	(typeof(T)==typeof(int))||
	(typeof(T)==typeof(long))||
	(typeof(T)==typeof(nint))||
	(typeof(T)==typeof(nuint))||
	(typeof(T)==typeof(sbyte))||
	(typeof(T)==typeof(float))||
	(typeof(T)==typeof(ushort))||
	(typeof(T)==typeof(uint))||
	(typeof(T)==typeof(ulong));
	}

.

[tannergooding]

@AntonLapounov, even if the type can't do anything, a user may still end up declaring a Vector<bool> or a Vector<bool>* themselves and the runtime needs to be robust in the face of that.

We could theoretically throw a TypeLoadException if we ever encounted an unsupported T, but I don't see much practical benefit in that, especially since we have expanded the set of T over time and there may be a Vector<T> declared on an unused code path (under the relevant IsSupported check).

[AntonLapounov]

Thank you for the explanation!