[LeadcodeDev]

Closes #162.

The defect

schemars does not emit #[serde(alias = ...)] — it writes only the canonical
name. Every aliased spelling was therefore declared invalid by rustmotion schema
while the engine accepts it perfectly, and the exported schema is the file a
generator reads to know what to write.

Proven on minimal documents, validated with jsonschema (Draft 7):

Document	before	after
{ "name": "float_3d" } in an animation	1 error	valid
{ "type": "progress_bar" }	1 error	valid

Two mechanisms

• Variant aliases widen the enum array they belong to: float_3d,
  progress_bar, flex-start/flex_start, flex-end/flex_end,
  space-between/-around/-evenly.
• Field aliases add a property mirroring the canonical one's schema.
  BorderRadius is the case the issue did not mention: its corners are declared
  as top-left and carry alias = "top_left", and the branch is
  deny_unknown_fields, so a scenario written the snake_case way was rejected
  outright.

Keyed by definition, not by value

Keying by value would have been shorter and wrong: another enum with a start
variant and no aliases would have inherited spellings the engine rejects — turning
a false negative into a false positive, which is the worse direction.
an_enum_that_shares_a_variant_name_but_not_its_aliases_is_left_alone pins it.

Drift is covered by reading the sources

every_serde_alias_in_the_sources_is_reachable_from_the_exported_schema scans
every .rs in the workspace for alias = "..." and fails naming any that the
exported schema does not contain. A new alias fails CI instead of quietly falling
out of the schema — the table cannot silently go stale.

What this does not fix

The issue attributes the examples' schema failures to this: "All trace back to
'float_3d' is not one of ['float3d']". Measured, that is not so — the examples
fail on 62 paths before and after, unchanged. Their causes are elsewhere:

• templating placeholders ($delay, $d) are not modelled as anything but
  numbers, so any templated scenario is invalid to the schema and fine to the
  engine — the same class of defect, a different cause;
• Draft-7 anyOf relevance reporting picks an arbitrary branch, so the messages
  name the wrong variant ('fade_in' is not one of ['fade_in_up']) and the real
  cause is unreadable.

Filed as its own issue rather than folded in here.

Gate

cargo fmt --all --check clean · cargo clippy --workspace --all-targets --features rustmotion/studio -D warnings clean · cargo test --workspace 1814 passed.