This modification is relevant for fixing regressions on recent changes
to the auto implicit behavior for inductive families.
The following declarations are now accepted:
```lean
inductive HasType : Fin n → Vector Ty n → Ty → Type where
| stop : HasType 0 (ty :: ctx) ty
| pop : HasType k ctx ty → HasType k.succ (u :: ctx) ty
inductive Sublist : List α → List α → Prop
| slnil : Sublist [] []
| cons l₁ l₂ a : Sublist l₁ l₂ → Sublist l₁ (a :: l₂)
| cons2 l₁ l₂ a : Sublist l₁ l₂ → Sublist (a :: l₁) (a :: l₂)
inductive Lst : Type u → Type u
| nil : Lst α
| cons : α → Lst α → Lst α
```
TODO: universe inference for `inductive` should be improved. The
current approach is not good enough when we have auto implicits.
TODO: allow implicit fixed indices that do not depend on indices that
cannot be moved to become parameters.
We go back to the original approach where we pattern matching
alternative variables as `FVar`s.
We fix the original problem we had by implementing a simple
unification procedure for alternative `FVar`s.
