One-of pattern matching not supported by static type analysis tools

According to the [Readme](https://github.com/danielgtaylor/python-betterproto/blob/master/README.md#one-of-support) it should be possible to access fields of a oneof-group by pattern matching so that static analysis tools can provide type hints:

```python3
test = Test()
match test:
case Test(on=value):
print(value) # value: bool
case Test(count=value):
print(value) # value: int
case Test(name=value):
print(value) # value: str
case _:
print("No value provided")
```

However, the tool `pyright` (used by the `pylance` extension for vscode) does not provide type hints for the second and third case. I first thought this was a bug in pyright, but [according to a pyright maintainer](https://github.com/microsoft/pyright/issues/8789#issuecomment-2299612496) this is actually intentional. Apparently, from the point of view of a type checker, the second and third case blocks are unreachable (this pattern even triggers a warning from pyright when configured accordingly).

## Possible solution

Let's assume we change the compiler to generate the following class:

```python3
@dataclass(eq=False, repr=False)
class Test(betterproto.Message):
on: Optional[bool] = betterproto.bool_field(1, group="foo")
count: Optional[int] = betterproto.int32_field(2, group="foo")
name: Optional[str] = betterproto.string_field(3, group="foo")
```

Then the following match statement is properly supported by pyright:

```python3
test = Test()
match test:
case Test(on=bool(value)):
print(value) # value: bool
case Test(count=int(value)):
print(value) # value: int
case Test(name=str(value)):
print(value) # value: str
case _:
print("No value provided")
```

I think one could argue that it is actually correct to mark these fields optional. What do you think?

## System Information
* python 3.11.9
* betterproto 2.0.0b7
* pylance v2024.8.1

Commencez par la section de prise en charge de one-of du README et la sortie du compilateur pour les champs oneof. Comparez les annotations Test générées avec les exemples de pattern matching de pyright, puis vérifiez que les champs oneof générés fournissent les indications de type attendues sans perturber le comportement oneof existant.

Rédigé par le modèle d'indexation à partir du texte de l'issue.