fix(jsonc-parser): strip UTF-8 BOM before parsing to fix Windows "InvalidSymbol at offset 0" errors

Windows editors often save UTF-8 files with a BOM (Byte Order Mark: \uFEFF).
When this is present, jsonc-parser reports InvalidSymbol at offset 0 because
the BOM is not valid JSON/JSONC syntax.

This commit strips the BOM before parsing, fixing issues #3164 where Windows
users report their opencode.jsonc file fails to parse even though it appears
to start with a valid '{' character.

Fixes: #3164
Co-authored-by: Jobdori <agent@yeongyu.kim>

src/shared/jsonc-parser.test.ts

@@ -139,6 +139,33 @@ describe("parseJsonc", () => {
     // then
     expect(() => parseJsonc(invalid)).toThrow()
   })
+
+  test("parses JSONC with UTF-8 BOM (Windows BOM files)", () => {
+    // given - JSON with UTF-8 BOM marker
+    const bom = "\uFEFF"
+    const jsonc = `${bom}{ "key": "value" }`
+
+    // when
+    const result = parseJsonc<{ key: string }>(jsonc)
+
+    // then
+    expect(result.key).toBe("value")
+  })
+
+  test("parses JSONC with BOM and comments", () => {
+    // given - JSONC with UTF-8 BOM and comments
+    const bom = "\uFEFF"
+    const jsonc = `${bom}{
+      // Windows editor saved with BOM
+      "key": "value"
+    }`
+
+    // when
+    const result = parseJsonc<{ key: string }>(jsonc)
+
+    // then
+    expect(result.key).toBe("value")
+  })
 })
 
 describe("parseJsoncSafe", () => {

src/shared/jsonc-parser.ts

@@ -10,6 +10,9 @@ export interface JsoncParseResult<T> {
 }
 
 export function parseJsonc<T = unknown>(content: string): T {
+  // Strip UTF-8 BOM if present (Windows UTF-8 with BOM files)
+  content = content.replace(/^\uFEFF/, "")
+
   const errors: ParseError[] = []
   const result = parse(content, errors, {
     allowTrailingComma: true,