fix(task-16): custom keys mirror AutoCAD's full AddCustomInfo rule, deletes are exempt, keys match case-insensitively

The 2b9791f fix mirrored only the characters the first review happened to
measure ('=' and ';'), so `custom={'AAA': 'first', 'A:B': 'x', 'ZZZ': 'never'}`
still raised from AddCustomInfo('A:B') on the live engine after subject and
AAA were applied -- the same half-write for 11 other keys. Re-measured on
AutoCAD 2026 by sweeping every character 0x20-0x7E through a scratch
document: exactly thirteen printable characters are 'Invalid key' anywhere
in a key (" * , / : ; < = > ? \ ` |); control characters, internal spaces,
a no-break space and unicode are accepted; leading/trailing whitespace
(space, tab, NBSP) is refused. `_CUSTOM_KEY_FORBIDDEN` is now that set,
the internal-NBSP refusal (wrong: measured accepted) is gone, and the
validator / tool docstrings state the measured rule instead of a closed
enumeration that was false.

A delete (`{key: None}`) is exempt from the AddCustomInfo syntax rules:
RemoveCustomByKey never validates syntax (measured: 'Key not found', never
'Invalid key'), and 2b9791f ran the check before the None branch, so a key
an ezdxf script wrote ('A=B', ' PAD ') showed in drawing_properties_get and
could never be removed. The line-break rule still applies both ways.

Also measured: custom keys are case-insensitive in AutoCAD (AddCustomInfo
over an existing case-variant is 'Duplicate key'; SetCustomByKey and
RemoveCustomByKey match either spelling and keep the stored one). The COM
routing was case-sensitive, so `{'PROJECT': ...}` over 'Project' went to
AddCustomInfo and failed mid-write -- the same class. Both engines now
match existing keys casefolded (update in place under the stored spelling,
never a second tag), and a request naming two case-variants is refused.

Tests: parametrized validator sweep over all thirteen characters, the
accepted set (control chars, NBSP, unicode, the other printables), the
delete exemption headlessly on a DXF carrying such keys and on the fake
SummaryInfo, case-variant refusal, and case-insensitive routing against a
fake that raises 'Duplicate key' / 'Key not found' like AutoCAD. Confirmed
live through the real ComBackend on a scratch document (Close(False)).

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FKFStjEFzpGyLq5bMpWxTd

backends/ezdxf_backend.py

@@ -6142,15 +6142,26 @@ class EzdxfBackend(AutoCADBackend):
                     "Save as R2004 or newer first."
                 )
             custom_vars = doc.header.custom_vars
+            # AutoCAD's custom keys are case-insensitive (measured on 2026:
+            # SetCustomByKey("PROJECT") updates an existing "Project" and keeps
+            # that spelling; a second AddCustomInfo is 'Duplicate key'), and
+            # ezdxf's CustomVars are plain case-sensitive tuples, so the match
+            # is done here: an existing tag is updated in place under its
+            # stored spelling, never duplicated under a second one.
+            props = custom_vars.properties
             for key, value in to_write.items():
-                if custom_vars.has_tag(key):
-                    custom_vars.replace(key, value)
+                folded = key.casefold()
+                hits = [i for i, (tag, _v) in enumerate(props) if tag.casefold() == folded]
+                if hits:
+                    props[hits[0]] = (props[hits[0]][0], value)
                 else:
                     custom_vars.append(key, value)
             deleted = []
             for key in to_delete:
-                if custom_vars.has_tag(key):
-                    custom_vars.remove(key, all=True)
+                folded = key.casefold()
+                kept = [(tag, v) for tag, v in props if tag.casefold() != folded]
+                if len(kept) != len(props):
+                    props[:] = kept
                     deleted.append(key)
             if to_write or deleted:
                 self._mark_dirty()