[core] Speed up NCZBLOCK reads and skip double AES for pre-decrypted NCZ (#4582)

- [x] I have read and followed the [Contribution Guidelines](https://git.eden-emu.dev/eden-emu/eden/src/branch/master/CONTRIBUTING.md#code-contributions).
- [x] I have read and followed the [AI Policy](https://git.eden-emu.dev/eden-emu/eden/src/branch/master/docs/policies/AI.md).
- [x] I have read and followed the [Coding Guidelines](https://git.eden-emu.dev/eden-emu/eden/src/branch/master/docs/policies/Coding.md) to the best of my ability.

-------------------
- Use a 32-slot direct-mapped block cache and reuse one ZSTD_DCtx per
  BlockReader instead of re-creating it per block.
- Find the covering section with std::lower_bound instead of a linear scan
  (helps BKTR updates with thousands of sections).
- Mark BlockNczFile as HasDecryptedSections() and drop its CTR pass; NCZ
  body is stored decrypted, so NcaFileSystemDriver now skips its AES layers
  for these files (no double decryption).
- Recognize .xcz as XCZ in AppLoader_XCI::IdentifyType.

thanks Reikatari

Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/4582
Reviewed-by: Maufeat <sahyno1996@gmail.com>

src/core/file_sys/vfs/vfs.h

@@ -112,6 +112,9 @@ public:
     // Returns whether or not the file can be read from.
     virtual bool IsReadable() const = 0;
 
+    // Returns true if this file serves NCA body data that is already decrypted (e.g. NCZ).
+    virtual bool HasDecryptedSections() const { return false; }
+
     // The primary method of reading from the file. Reads length bytes into data starting at offset
     // into file. Returns number of bytes successfully read.
     virtual std::size_t Read(u8* data, std::size_t length, std::size_t offset = 0) const = 0;