Interval: make public fields final

[stevenschlansker]

The class docs already state that Interval is immutable, but this is not enforced.
Followup to #4901

I tried to fix up existing sites where this supposedly immutable class is mutated, to instead replace with a new value.

[stevenschlansker]

I think this variable oldb is redundant since we already saved I.b above at line 656. Let me know if you'd like me to remove this. (I didn't, in the spirit of keeping the change minimal)

[ericvergnaud]

Only test coverage would tell us that. Let's not touch this.

[ericvergnaud]

Only test coverage would tell us that. Let's not touch this.

[ericvergnaud]

@parrt blessed.
I'm very supportive of this change. The class comment is as follows:
/** An immutable inclusive interval a..b */
Note the immutable in there, which expresses the original intent.
The fact that we violated it internally might have been justified by performance, memory, whatever... I suspect mutations of intervals in the cache is the root cause of issue #4901, more than the cache itself. There doesn't seem to be any code checking whether the interval being mutated is in the interval cache, and should therefore absolutely not be mutated. This change ensures that the behavior complies with the intent.
It probably also helps HotSpot do a better job since it can cache a and b.