Exercise 2: What the Plan-as-PR-Comment Pattern Shows Beyond the .tf Diff — Possible Solution
====================================================================

A raw `.tf` file diff shows only the TEXTUAL change to the
configuration -- which lines were added, removed, or edited. It says
nothing about the ACTUAL, real-world CONSEQUENCE of that textual
change, because Terraform's behavior in response to a given edit isn't
always obvious just from reading it.

The chapter's own example makes this concrete: "a one-line variable
change can produce a plan that recreates a dozen resources." Someone
reading only the diff sees a single changed line -- say, a variable's
default value edited, or a `count`/`for_each` collection modified in a
way that shifts positional indices (exactly Chapter 3 Exercise 2's own
scenario). Nothing in that one-line diff visually signals "this is
about to destroy and recreate 12 real resources" -- that consequence
only becomes visible by actually running Terraform's own plan logic
against the real dependency graph and real current state.

Posting the ACTUAL `plan` output as a PR comment surfaces exactly that
hidden consequence, in Terraform's own `+`/`-`/`~` notation, before
anyone approves the merge -- the reviewer sees precisely what will be
created, destroyed, or changed in real infrastructure, not just what
text changed in a file. This lets a reviewer catch a genuinely
dangerous change (an unintended mass-destroy-and-recreate, say) that
would have been completely invisible from the diff alone, and would
otherwise only have been discovered after `apply` had already run.

WHY THIS WORKS AS AN ANSWER
------------------------------
This uses the chapter's own concrete example (a one-line change causing
a dozen recreations) to show specifically why a diff alone can't reveal
Terraform's actual planned consequence, rather than a generic "plan is
more informative than a diff" claim.