*: s/rng/rep/ for vars of type Replica

[jordanlewis]

It's confusing to have variables named rng when their type is Replica and not Range or similar. This is a large cosmetic change that fixes a lot of those occurrences.

---

This change is

[tamird]

LGTM without reviewing

[RaduBerinde]

LGTM

Nit: I've usually seen repl for Replica vars.. But feel free to ignore, unless you want to try to do it automatically (e.g generate a diff, s/\<rep\>/repl/g in the patch, then reapply).

[jordanlewis]

There unfortunately seems to be a mix of both, but repl wins out by a bit so I've gone ahead and taken your suggestion.

[21:29]% ag " repl[ \.]" | wc -l
84
[21:30]% ag " rep[ \.]" | wc -l
69

[petermattis]

We also use r for *Replica vars. And that is the receiver name for all of the methods on *Replica. I personally prefer rep for such vars, but that preference is not strong.

[RaduBerinde]

---

Review status: 0 of 16 files reviewed at latest revision, all discussions resolved, all commit checks successful.

---

Comments from Reviewable

[tbg]

LGTM but please don't merge before #10327.

[jordanlewis]

@petermattis r seems a bit terse outside of the method receiver context. I hear your preference for rep but it does seem like there are already more instances of repl - considering the inconsistency here I'm just going to go ahead and merge this, since either one is strictly better than the incorrect rng naming.