Small speedup for dataclass __eq__ and __repr__

[rhettinger]

1. Have__eq__ do a series of and comparisons rather than building a tuple and relying on tuple.__eq__.
2. Let __repr__ put the type name in the f-string rather than calling string concatenation explicitly.

Old code:

 def __eq__(self,other): if other.__class__ is self.__class__: return (self.a,self.b,)==(other.a,other.b,) def __repr__(self): return self.__class__.__qualname__ + f"(a={self.a!r}, b={self.b!r})" 

New code:

 def __eq__(self,other): if other.__class__ is self.__class__: return self.a==other.a and self.b==other.b def __repr__(self): return f"{self.__class__.__qualname__}(a={self.a!r}, b={self.b!r})" 

Timings for a dataclass with two integer fields:

 Both Equal First Equal Neither Equal Repr ========== =========== ============= ============ Baseline 0.09769874 0.10188975 0.10144600 0.18586733 With PR 0.08537712 0.08647508 0.070007080 0.18308112 

[carljm]

Looks good!

[ericvsmith]

Those look great, @rhettinger! Thanks. I'll do a detailed code review then commit it.

[hroncok]

I belive I found a regression: #116647

[terryjreedy]

Or you found that this resulted in a perhaps inadvertent bugfix ;-). Depends on Eric's intention for NANs in dataclasses.