✨(backend) add mimetype detection logic from drive

Reused the logic developed by the team working on drive.
This is usefull for our own upload file backend (that will
come in later commits).

Dockefile was updated to add a required system dependency.

We might want to put this shared logic in a lib.

src/backend/core/utils.py

@@ -7,6 +7,8 @@ Utils functions used in the core app
 
 import hashlib
 import json
+import logging
+import mimetypes
 import random
 import secrets
 import string
@@ -18,6 +20,7 @@ from django.core.files.storage import default_storage
 
 import aiohttp
 import botocore
+import magic
 from asgiref.sync import async_to_sync
 from livekit.api import (  # pylint: disable=E0611
     AccessToken,
@@ -29,6 +32,8 @@ from livekit.api import (  # pylint: disable=E0611
     VideoGrants,
 )
 
+logger = logging.getLogger(__name__)
+
 
 def generate_color(identity: str) -> str:
     """Generates a consistent HSL color based on a given identity string.
@@ -344,3 +349,65 @@ def generate_room_slug():
         for size in sizes
     ]
     return "-".join(parts)
+
+
+def detect_mimetype(file_buffer: bytes, filename: str | None = None) -> str:
+    """
+    Detect MIME type using multiple methods for better accuracy.
+
+    This function combines:
+    1. Magic bytes detection (python-magic) - most reliable for actual file content
+    2. File extension detection (mimetypes) - useful as fallback or for validation
+
+    Args:
+        file_buffer: The file content buffer (first bytes of the file)
+        filename: Optional filename to extract extension from
+
+    Returns:
+        str: The detected MIME type
+
+    Notes:
+        Originally from https://github.com/suitenumerique/drive/blob/564822d31f071c6dfacd112ef4b7146c73077cd9/src/backend/core/api/utils.py#L166 # pylint:disable=line-too-long
+    """
+    # Initialize magic detector
+    mime_detector = magic.Magic(mime=True)
+
+    # Method 1: Detect from file content (magic bytes) - most reliable
+    mimetype_from_content = mime_detector.from_buffer(file_buffer)
+
+    # If we have a filename, try extension-based detection as well
+    mimetype_from_extension = None
+    if filename:
+        # Use mimetypes module to guess from extension
+        # Use guess_file_type (Python 3.13+) instead of deprecated guess_type
+        mimetype_from_extension, _ = mimetypes.guess_file_type(filename, strict=False)
+
+    logger.info("detect_mimetype: mimetype_from_content: %s", mimetype_from_content)
+    logger.info("detect_mimetype: mimetype_from_extension: %s", mimetype_from_extension)
+
+    # Strategy: Prefer content-based detection, but use extension if:
+    # 1. Content detection returns generic types (application/octet-stream, text/plain)
+    # 2. Content detection fails or returns None
+    # 3. Extension detection provides a more specific type
+
+    # Generic/unreliable MIME types that we should try to improve
+    generic_types = {
+        "application/octet-stream",
+        "application/x-ole-storage",  # used by .xls, .doc and .ppt
+        "application/zip",
+        "text/plain",
+    }
+
+    # If content detection gives us a generic type and we have extension info
+    if mimetype_from_content in generic_types and mimetype_from_extension:
+        # Use extension-based detection if it's more specific
+        if mimetype_from_extension not in generic_types:
+            return mimetype_from_extension
+
+    # If content detection failed, returned None or is a generic type, use extension if available
+    if not mimetype_from_content or mimetype_from_content in generic_types:
+        if mimetype_from_extension:
+            return mimetype_from_extension
+
+    # Default to content-based detection (most reliable)
+    return mimetype_from_content or "application/octet-stream"