Page type-check: strict throughout, a typed $, null checks (#236)

$<T> returns the element type it is given and assumes the element is there: 122 lookups take
their type from the tag their id is on in the markup. A handler reads its target through a
cast, since a typed element will not take a handler whose event names a narrower target.
strictNullChecks found no bug the page had; its 94 places were values the code already handles
in ways the type-check could not follow (data- attributes the page writes, a map entry set the
line before, Share drawn only for an item with a link), each now said so.

isPlayable stays a plain boolean: as a type guard its false would have meant "no file", and it
is also false for a file that is not audio or video.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

CLAUDE.md

@@ -154,9 +154,11 @@ node tests/page-smoke.js
 node --test tests/page-unit.js
 ```
 
-The first type-checks `web/src`: every parameter and variable has a type, and what the API sends
-is typed in `web/src/types.ts`, which mirrors the Rust structs and changes with them. Not yet
-null-safe: `strictNullChecks` is off and `$` returns `any` (#236). The second
+The first type-checks `web/src` in TypeScript's `strict` mode: every parameter and variable has a
+type, what the API sends is typed in `web/src/types.ts`, which mirrors the Rust structs and changes
+with them, and a value that may be null has to be checked before use. `$<T>` returns the element
+type it is given (`$<HTMLInputElement>('#nurl')`) and assumes the element is there; one that may
+not be is written `$(...)?.`, and the page check below proves the ones wired at load exist. The second
 builds the page as shipped and runs its script against a stub DOM, checking every selector it
 wires at load actually exists. That check exists because a patch once anchored on a deleted
 function, `String.replace` silently matched nothing, and the whole UI died with a