fix(vocab): dedupe bulk import case-insensitively within the paste

Previously the bulk import ran new Set(...) on raw trimmed strings
before lowercasing, so 'ACME' and 'acme' both survived the first
dedupe pass. Neither existed in the store, so both got added —
defeating the commit message's claim that pasting the same block
twice with different casing is a no-op.

Collapse case variants at the initial dedupe step using a lowercase
seen-set, keeping the first occurrence's casing as written.

Co-authored-by: jars <jakejars@users.noreply.github.com>

src/lib/pages/SettingsPage.svelte

@@ -196,15 +196,19 @@
   async function addBulkVocabTerms() {
     const raw = bulkVocabText;
     // Accept newline-separated OR comma-separated (or mixed) — whichever the
-    // user pasted. Trim each entry, drop empties, dedupe within the input.
+    // user pasted. Trim each entry, drop empties, and dedupe case-insensitively
-    const candidates = Array.from(
+    // so "ACME" and "acme" in the same paste collapse to a single term
-      new Set(
+    // (Whisper prompts don't care about case).
-        raw
+    const seen = new Set();
-          .split(/[\n,]+/)
+    const candidates = [];
-          .map((entry) => entry.trim())
+    for (const entry of raw.split(/[\n,]+/)) {
-          .filter((entry) => entry.length > 0),
+      const trimmed = entry.trim();
-      ),
+      if (!trimmed) continue;
-    );
+      const key = trimmed.toLowerCase();
+      if (seen.has(key)) continue;
+      seen.add(key);
+      candidates.push(trimmed);
+    }
     if (candidates.length === 0) {
       vocabularyError = "Nothing to import — paste one term per line or separated by commas.";
       return;