Get rid of equality override on array types

### Summary
`ndarray` equality returns an array of boolean values, checking equality element-wise, by default. This should be preserved in our array variable value types, leaving it up to the client to use that as they need.

**Example**
```python
import numpy as np

x = acvi.IntegerArrayValue(values=[1, 2])
y = acvi.IntegerArrayValue(values=[1, 2])
z = acvi.IntegerArrayValue(values=[1, 3])

# default equality behavior
x_eq_y = (x == y) # [True, True]: acvi.IntegerArrayValue
x_eq_z = (x == z) # [True, False]: acvi.IntegerArrayValue

# client could then do this to check arrays are overall equal
all_x_eq_y = np.all(x == y) # True: np.bool_
all_x_eq_z = np.all(x == z) # False: np.bool_
```

### Tasks
- [ ] Remove `__eq__()` methods from (Real, Integer, Boolean, String)ArrayValue
- [ ] Fix unit tests accordingly

Locate the RealArrayValue, IntegerArrayValue, BooleanArrayValue, and StringArrayValue definitions and their __eq__() methods, then inspect the unit tests covering array equality. Run those tests before and after removing the overrides, and update expectations so equality returns element-wise array values while the full test suite passes.

Written by the indexing model from the issue text.