Fix casts in overflow checking function

[CaelmBleidd]

Fix the wrong way to cast primitive values in the function related to overflowing checks

Description

Added an explicit cast for arguments of the check function. They were wrapped into a primitive value that didn't change their symbolic type. That led to an error where we process primitive values of different sorts.

Fixes #2022

How to test

Automated tests

org.utbot.examples.math.OverflowAsErrorTest#testByteWithIntOverflow

Manual tests

I tested the example from the issue and a similar one. Note that an example from the issue doesn't contain overflow since + casts both operands to an int type, therefore, since the second one is a constant and the first one is a byte value, an integer overflow can never occur. If we transform the second value into a symbolic value, we'll find an overflow error.

Self-check list

• I've set the proper labels for my PR (at least, for category and component).
• PR title and description are clear and intelligible.
• I've added enough comments to my code, particularly in hard-to-understand areas.
• The functionality I've repaired, changed or added is covered with automated tests.
• Manual tests have been provided optionally.
• The documentation for the functionality I've been working on is up-to-date.

[Markoutte]

I think we need discuss more about this feature, because, it can have strange behavior from user point of view. For example, we have this test be reported as it has overflow, but in real life it does not:

    public static int foo(byte a, short b) {
        return a + b;
    }

Despite this, I think this fix should be merged as is, because it fixes critical problem.

[CaelmBleidd]

I think we need discuss more about this feature, because, it can have strange behavior from user point of view

Agree, created an issue for it