Refine all `catch` blocks to use consistent naming for `Exception` variables

[jxblum]

This task consistently aligns all of Spring Data Redis's catch blocks / Exception handlers with the core Spring Framework's Exception (or Throwable) variable reference names, which is primarily (and nearly consistently)... ex.

Additionally, the core Spring Framework also uses ignore in places where the Exception (or Throwable) is ignored.

NOTE: Unfortunately, only in a couple/three places (or so) did the core Spring Framework use inconsistent Exception variable names, such as exc. However, nearly all are named ex.

Now, in Spring Data Redis, ex and ignore are consistently and appropriately used everywhere. The are only 2 exceptions to this convention:

1. If the Exception handler is in a TEST class and "expected" , but not referenced or used in any way, then the "expected" keyword was used (only in TESTS).

2. In only 1 place (i.e. ScanCursor) was the Exception variable named "nested" due to try-catch block nesting, as in:

try {
    // Exceptional code
} catch (Exception ex) {
    try {
        // More Exceptional code
    } catch (RuntimeException nested) {
        ex.addSuppressed(nested);
    }
}

In a few places, the Exception handler logic in catch blocks were refactored to be more readable, separating Exception message construction/formatting from the instantiation and throwing of the Exception itself.

Finally, String.format(..) was consistently applied to all Exception messages that require formatting.

I used the following REGEX expressions to search and modify our Exception blocks:

catch \(.*Exception e\)
catch \(.*Exception ex\)
catch \(.*Exception expected\)
catch \(.*Exception ignore\)
catch \(.*Exception [^ex|^ignore|^nested].*\)