feat(cli): colitur check and colitur new-overlay

Writing a local calendar had no feedback loop. An overlay is applied, not
validated -- that stays true, and the five test layers still cannot vouch for
a user's file -- but before this the only way to learn whether yours did what
you meant was to generate a year of output and grep for your own slug, and the
only way to see that a directive matched nothing was to notice a warning
scroll past among 365 lines.

`check` loads each overlay, applies it to the real shipped calendar, and
reports the directive counts, the slug each one targets, and any directive
that found no target. It exits 2 when a file fails to load or a directive
matched nothing, so it composes into a Makefile or a pre-commit hook rather
than merely being readable. It is applied to the SHIPPED calendar and not to
an empty layer on purpose: against an empty one every Suppress would fail
trivially and the check would be worthless.

It answers three narrow questions -- does the file parse, does every directive
find its target, what does the merged result contain. It does not validate a
calendar against the rubrics and cannot, and both the help text and the man
page say so rather than letting the name imply more than it does.

`new-overlay` prints a starter to stdout for redirection, rather than writing
a file where it likes. Every value in it is a placeholder that will appear in
`day` output if left unedited, so a half-finished overlay is visible rather
than silently inert, and it documents the three date shapes and the legal
values for each closed field inline.

load_ef_layer now returns its diagnostics instead of printing them: day and
readings still want them on stderr beside a year of output, while check wants
them on stdout, attributed to the overlay that produced them, and counted.
Printing at the source made the second impossible.

The cram test round-trips new-overlay through check rather than pinning the
template line by line -- editing its prose should not fail a test, but a
syntax error in it still must.

1 files changed, 35 insertions, 3 deletions

diff --git a/test/cli.t b/test/cli.t
index 0c79db4..d735e69 100644
--- a/test/cli.t
+++ b/test/cli.t
@@ -17,7 +17,7 @@ A year outside the supported domain is rejected (exit 2):
 No/garbage arguments give a usage error (exit 2):
 
   $ colitur
-  colitur: usage: colitur easter <year> | colitur temporal <year> | colitur day <year> | colitur readings <year>  (try: colitur --help)
+  colitur: usage: colitur easter <year> | colitur temporal <year> | colitur day <year> | colitur readings <year> | colitur check FILE | colitur new-overlay  (try: colitur --help)
   [2]
 
 The EF temporal cycle for a year, one line per day:
@@ -327,17 +327,49 @@ A flag needing a value, given none:
 
   $ colitur day 2026 --overlay
   colitur: --overlay needs a file path
-  colitur: usage: colitur easter <year> | colitur temporal <year> | colitur day <year> | colitur readings <year>  (try: colitur --help)
+  colitur: usage: colitur easter <year> | colitur temporal <year> | colitur day <year> | colitur readings <year> | colitur check FILE | colitur new-overlay  (try: colitur --help)
   [2]
 
 An unknown option is rejected rather than treated as a positional word:
 
   $ colitur day 2026 --diocese
   colitur: unknown option --diocese
-  colitur: usage: colitur easter <year> | colitur temporal <year> | colitur day <year> | colitur readings <year>  (try: colitur --help)
+  colitur: usage: colitur easter <year> | colitur temporal <year> | colitur day <year> | colitur readings <year> | colitur check FILE | colitur new-overlay  (try: colitur --help)
   [2]
 
 The shipped example overlay is runnable documentation, and it must actually
 load -- an example that silently rotted would be worse than none. The cram
 sandbox cannot reach data/, so the assertion that it loads and applies lives
 in test_lectionary_ef.ml, which reads it from the source tree directly.
+
+`new-overlay` prints a starter file to stdout for redirection, and what it
+prints must itself be valid -- a template that does not load is worse than no
+template, because it teaches the wrong shape. Round-tripped here rather than
+pinned line by line, so editing the template's prose does not fail this test
+while a syntax error in it still does:
+
+  $ colitur new-overlay > starter.sexp
+  $ colitur check starter.sexp
+  starter.sexp: ok -- overlay my-parish, 2 directive(s): 2 add, 0 suppress, 0 replace, 0 edit
+    every directive found its target
+    add      my-local-patron
+    add      my-dedication
+
+`check` exits 2 when a directive matches nothing, so it is usable in a
+Makefile or a pre-commit hook, not merely readable:
+
+  $ printf '((id p) (directives ((Suppress no-such-slug))))' > bad.sexp
+  $ colitur check bad.sexp
+  bad.sexp: ok -- overlay p, 1 directive(s): 0 add, 1 suppress, 0 replace, 0 edit
+    MATCHED NOTHING: overlay p: suppress no-such-slug: slug not present; nothing to suppress
+    suppress no-such-slug
+  [2]
+
+`citations` and `layer` may be omitted from an added celebration; they default
+to empty and to the overlay's own id:
+
+  $ printf '((id tiny) (directives ((Add ((date (Fixed (month 5) (day 20))) (cel ((slug tiny-feast) (names ((en "Tiny"))) (rank Class3) (status Feast) (colour White) (subject Saint))))))))' > tiny.sexp
+  $ colitur check tiny.sexp
+  tiny.sexp: ok -- overlay tiny, 1 directive(s): 1 add, 0 suppress, 0 replace, 0 edit
+    every directive found its target
+    add      tiny-feast