Move outstanding work from TODO.md into the issue tracker

TODO.md carried two records in one file: what had been done, with the
measurements behind it, and what was left. The second half is what a tracker
is for, and keeping it here has already cost something -- AGENTS.md records a
round where eleven entries described code that had already changed, and this
file admitted to three more.

Every open item is now an issue on source.starfort.tech/andrew/libakgl,
labelled by kind and blast radius and milestoned by what it can land in: 0.9.x
for anything that breaks no ABI, 0.10.0 for new or changed public symbols,
1.0.0 for the design work. Four are epics: the performance plan (#60),
coverage (#61), actor rotation (#62), and the false header comments (#63).

Verified against the tree before filing rather than transcribed. Three entries
were already fixed and were not filed: the akgl_path_relative context leak, the
akgl_draw_background test extension, and the SDL enumeration audit -- keyboards,
gamepads and mappings are all freed in CLEANUP today. Two were reworded because
the code had moved: the fonts item is a missing teardown entry point rather than
a missing API, since akgl_text_unloadallfonts exists, and draw_world's tilemap
call is already bounded by numlayers, so only the per-layer actor rescan remains.

TODO.md keeps the part a tracker has no place for: why a decision went the way
it did, what the measurement was, and which arguments turned out to be wrong.

TODO.txt is deleted. Four of its eight entries had shipped -- actor-to-actor
collision, actor-to-world collision, automatic facing, image layers -- and the
four that had not are #74 through #77, with the GPU renderer's research links
kept because that is the part that took the time.

Every reference that named an item number or a moved section is repointed, in
the manual, the headers, the tests and the examples.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-Authored-By: Andrew Kesterson <andrew@aklabs.net>

docs/19-utilities.md

@@ -161,8 +161,8 @@ Three things to know:
 
 > **Stale defect note.** `util.h:124-128` warns that the fallback path "returns straight out
 > of the ENOENT handler and so never releases the error context it was handling", costing one
-> of libakerror's 128 context slots per call for the life of the process, and `TODO.md` item
-> 15 records the same thing as open. **Both are out of date.** `src/util.c:115-129` sets a
+> of libakerror's 128 context slots per call for the life of the process, and `TODO.md`
+> recorded the same thing as open. **Both were out of date.** `src/util.c:115-129` sets a
 > flag in the `HANDLE(errctx, ENOENT)` block and calls `path_relative_root` *after* `FINISH`,
 > which is exactly the fix that entry proposes, and the code carries a comment explaining
 > why. The leak is gone; the two notes describing it are not.
@@ -332,7 +332,7 @@ for an object member.
 > `HANDLE_GROUP(e, AKERR_OUTOFBOUNDS)` arm, placed *above* the arm holding the `memcpy`
 > because `HANDLE_GROUP` emits no `break` and every arm falls into that body, and
 > `tests/json_helpers.c` covers it through both the integer and object index accessors.
-> `TODO.md` item 18 records it as fixed. The header comment wants correcting in its own
+> `TODO.md` records it as fixed. The header comment wants correcting in its own
 > commit.
 
 `defsize` is trusted, not derived: it is a `memcpy` through `void *` with no type