Coverage measures the page's Node tests and runs the Rust on Postgres (#407)

page-unit.ts and native-bridge.ts run the page's script outside a
browser, so the page's coverage counted none of what they test, and
native.ts read as mostly untested. In a coverage run they now run the
script from a file of its own, named as the browser names it, under
NODE_V8_COVERAGE (scripts/coverage-node.sh, before the browser suite in
npm run coverage:page and coverage.sh api), and the report's teardown
takes their record in, mapped to web/src by the build's source map: the
page's lines went from 73.3% to 80.0% of 3,043, native.ts to 93 of 120.

The cargo tests in coverage.sh's container run on Postgres, as
production does, when IPX_TEST_DATABASE_URL names the _test database,
from the shell or else /src/.envrc (that variable only); on SQLite
without it. The integration tests' own files are left out of the
totals; a crate's inline mod tests cannot be on stable Rust.

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

tests/page-unit.ts

@@ -9,13 +9,13 @@
 // a second <script> on a page would, so `run` evaluates an expression there.
 import { test } from 'node:test';
 import assert from 'node:assert/strict';
-import { makeContext, vm } from './dom-stub.ts';
+import { makeContext, vm, scriptName } from './dom-stub.ts';
 import { buildPage } from '../web/build.mts';
 
 const { html, js } = buildPage('index.html');
 const { ctx } = makeContext(html, js);
 vm.createContext(ctx);
-vm.runInContext(js, ctx, { filename: 'index.html<script>', timeout: 5000 });
+vm.runInContext(js, ctx, { filename: scriptName(js, 'index.html<script>'), timeout: 5000 });
 // What the page's script answers, from inside the VM: its types are the script's, not this file's.
 const run = (expr: string): any => vm.runInContext(expr, ctx);
 // Time as the page sees it, for the helpers that count from now.