Document optional line numbers

Chapter 2 gets the rule and the refusal, chapter 4 gets the payoff for
LABEL, chapter 9 says DSAVE writes the numbers it handed out and to
RENUMBER first if you want gaps, chapter 10 shows a host loading numberless
source, chapter 13 gets the QuickBASIC-shaped divergence, and chapter 14's
source[] passage gets its second half.

Chapter 14 said "two prescans" and listed two; there were three before this
and there are four now, so it lists all four and says which of them reports
against the right line.

TODO.md section 6 records four things found on the way and deliberately not
fixed: set_label() filing into the active scope rather than the root, three
prescans reporting the wrong line number, duplicate written line numbers
still replacing silently, and renumber.c's file-scope scratch arrays.

examples/embed.c runs the same program twice, numbered and not, so the
example compiles the feature rather than describing it. Its header pointed
at ./build/examples/embed, which is not where the binary lands.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

docs/10-embedding.md

@@ -56,6 +56,28 @@ akerr_ErrorContext AKERR_NOIGNORE *run_script(const char *source)
 `akbasic_runtime_run(rt, n)` runs at most `n` steps and returns. That bound is what
 keeps a runaway script from owning your process.
 
+**The source you hand `akbasic_runtime_load()` does not need line numbers.** A game
+script is written in a text editor and branches by `LABEL`, so numbering its lines is
+work with nothing on the other end of it:
+
+```c wrap=hostloop
+static const char *SCRIPT =
+    "PRINT \"SPAWNING\"\n"
+    "FOR I# = 1 TO HEALTH#\n"
+    "  PRINT I#\n"
+    "NEXT I#\n"
+    "GOTO DONE\n"
+    "PRINT \"NOT REACHED\"\n"
+    "LABEL DONE\n"
+    "PRINT \"READY\"\n";
+```
+
+Lines that carry a number are filed under it and lines that do not are given the next
+one going, so the two can be mixed and a numbered script still loads exactly as it did.
+What a script may *not* do is `GOTO 100` when nothing wrote a line 100 — the interpreter
+refuses that before the first line runs rather than branching somewhere plausible and
+wrong.
+
 `akbasic_runtime_settime()` is how the interpreter knows what time it is. It reads no
 clock of its own, because it owns no loop. If you never call it, every duration expires
 immediately — audible, but never a hang.