fix(api): optimize HEAD requests for chapter downloads (#1601)

Previously, handling a HEAD request on the chapter download endpoint was inefficient as it triggered the full CBZ file generation process in-memory just to retrieve metadata like Content-Length and Content-Disposition. This caused unnecessary latency especially for OPDS clients.

This commit introduces a separate, lightweight path for HEAD requests.

- A new `getCbzMetadataForDownload` method is added to `ChapterDownloadHelper` to calculate the filename and file size without generating an archive stream.
- The `ChaptersFilesProvider` interface is updated with a `getArchiveSize()` method, implemented by both `ArchiveProvider` and `FolderProvider`, to retrieve the total size.
- The `MangaController` now differentiates between GET and HEAD methods, invoking the appropriate helper to ensure HEAD requests are served instantly with only the required metadata.

server/src/main/kotlin/suwayomi/tachidesk/manga/impl/ChapterDownloadHelper.kt

@@ -94,4 +94,26 @@ object ChapterDownloadHelper {
 
         return Triple(cbzFile.first, fileName, cbzFile.second)
     }
+
+    fun getCbzMetadataForDownload(chapterId: Int): Triple<String, Long, String> { // fileName, fileSize, contentType
+        val (chapterData, mangaTitle) =
+            transaction {
+                val row =
+                    (ChapterTable innerJoin MangaTable)
+                        .select(ChapterTable.columns + MangaTable.columns)
+                        .where { ChapterTable.id eq chapterId }
+                        .firstOrNull() ?: throw IllegalArgumentException("ChapterId $chapterId not found")
+                val chapter = ChapterTable.toDataClass(row)
+                val title = row[MangaTable.title]
+                Pair(chapter, title)
+            }
+
+        val scanlatorPart = chapterData.scanlator?.let { "[$it] " } ?: ""
+        val fileName = "$mangaTitle - $scanlatorPart${chapterData.name}.cbz"
+
+        val fileSize = provider(chapterData.mangaId, chapterData.id).getArchiveSize()
+        val contentType = "application/vnd.comicbook+zip"
+
+        return Triple(fileName, fileSize, contentType)
+    }
 }

server/src/main/kotlin/suwayomi/tachidesk/manga/impl/download/fileProvider/ChaptersFilesProvider.kt

@@ -220,6 +220,8 @@ abstract class ChaptersFilesProvider<Type : FileType>(
 
     abstract fun getAsArchiveStream(): Pair<InputStream, Long>
 
+    abstract fun getArchiveSize(): Long
+
     private fun maybeConvertPages(chapterCacheFolder: File) {
         val conversions = serverConfig.downloadConversions.value
 

server/src/main/kotlin/suwayomi/tachidesk/manga/impl/download/fileProvider/impl/ArchiveProvider.kt

@@ -108,6 +108,11 @@ class ArchiveProvider(
         return cbzFile.inputStream() to cbzFile.length()
     }
 
+    override fun getArchiveSize(): Long {
+        val cbzFile = File(getChapterCbzPath(mangaId, chapterId))
+        return if (cbzFile.exists()) cbzFile.length() else 0L
+    }
+
     private fun extractCbzFile(
         cbzFile: File,
         chapterFolder: File,

server/src/main/kotlin/suwayomi/tachidesk/manga/impl/download/fileProvider/impl/FolderProvider.kt

@@ -101,4 +101,11 @@ class FolderProvider(
         val zipData = byteArrayOutputStream.toByteArray()
         return ByteArrayInputStream(zipData) to zipData.size.toLong()
     }
+
+    override fun getArchiveSize(): Long {
+        val chapterDir = File(getChapterDownloadPath(mangaId, chapterId))
+        if (!chapterDir.exists() || !chapterDir.isDirectory) return 0L
+        // Approximation: actual CBZ size is slightly larger due to ZIP metadata, but sufficient for Content-Length header.
+        return chapterDir.listFiles()?.filter { it.isFile }?.sumOf { it.length() } ?: 0L
+    }
 }