feat(ui): add /cv route, site-wide lowercase, no-cookies footer

reproduces the legacy cv (previously at grenade.github.io/cv) as a
react-router /cv route, fetched at runtime from the same gist. moves
the lowercase aesthetic from per-element overrides to a single body-
level rule so a future toggle can flip it from one place. adds a small
site-wide footer noting why no cookie consent banner is shown.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

ui/src/lib/cvDates.ts

@@ -0,0 +1,91 @@
+// Normalizes the date interval line of a CV entry. The legacy implementation
+// at cv/src/App.js:139 chained .replace() calls per month name; this collapses
+// that into a single regex pass.
+
+const MONTH_ABBREV: Record<string, string> = {
+  january: 'jan',
+  february: 'feb',
+  march: 'mar',
+  april: 'apr',
+  // may stays "may"
+  june: 'jun',
+  july: 'jul',
+  august: 'aug',
+  september: 'sep',
+  october: 'oct',
+  november: 'nov',
+  december: 'dec',
+};
+
+const MONTH_RE = new RegExp(
+  `\\b(${Object.keys(MONTH_ABBREV).join('|')})\\b`,
+  'gi',
+);
+
+// Strip leading/trailing markdown # / whitespace, abbreviate months. Casing
+// is left to the body-level `text-transform: lowercase` so a future toggle can
+// flip it from a single place.
+export function normalizeInterval(line: string): string {
+  return line
+    .replace(/^[\s#]+|[\s#]+$/g, '')
+    .replace(MONTH_RE, (m) => MONTH_ABBREV[m.toLowerCase()] ?? m);
+}
+
+// Strip markdown link syntax (e.g. "[text](url)") down to just the text.
+export function stripMdLinks(s: string): string {
+  return s.replace(/\[([^\]]+)\]\([^)]+\)/g, '$1');
+}
+
+// Parse the title / location-role / interval triple out of a CV entry's
+// markdown content. If the first line embeds a PNG URL (icon-style entry),
+// indices shift down by one.
+export interface ParsedHeader {
+  // Plain text, markdown links stripped — for alt= attributes and similar.
+  title: string;
+  locationRole: string;
+  // Markdown source with leading #s/whitespace stripped — for inline rendering
+  // so [text](url) links render as proper anchors.
+  titleMd: string;
+  locationRoleMd: string;
+  iconUrl: string | null;
+  interval: string;
+}
+
+export function parseEntryHeader(content: string): ParsedHeader {
+  const lines = content.split('\n');
+  const firstLine = lines[0] ?? '';
+  const hasIcon = firstLine.includes('.png');
+  const iconUrl = hasIcon
+    ? (firstLine.match(/https:[^ )\]]+\.png/)?.[0] ?? null)
+    : null;
+
+  const titleLine = hasIcon ? (lines[1] ?? '') : firstLine;
+  const locRoleLine = hasIcon ? (lines[2] ?? '') : (lines[1] ?? '');
+  const intervalLine = hasIcon ? (lines[3] ?? '') : (lines[2] ?? '');
+
+  const titleMd = titleLine.replace(/^[\s#]+|[\s#]+$/g, '');
+  const locationRoleMd = locRoleLine.replace(/^[\s#]+|[\s#]+$/g, '');
+
+  return {
+    title: stripMdLinks(titleMd),
+    locationRole: stripMdLinks(locationRoleMd),
+    titleMd,
+    locationRoleMd,
+    iconUrl,
+    interval: normalizeInterval(intervalLine),
+  };
+}
+
+// Anchor id for an entry: combines the section name and a slug of the title
+// line. Mirrors the legacy id format at cv/src/App.js:71.
+export function entryAnchorId(sectionName: string, content: string): string {
+  const lines = content.split('\n');
+  const firstLine = lines[0] ?? '';
+  const titleLine = firstLine.includes('.png') ? (lines[1] ?? '') : firstLine;
+  const slug = stripMdLinks(titleLine.replace(/^[\s#]+|[\s#]+$/g, ''))
+    .toLowerCase()
+    .replace(/[^a-z0-9\s-]/g, '')
+    .replace(/\s+/g, '-')
+    .replace(/-+/g, '-');
+  return `${sectionName}-${slug}`;
+}