f-string regression in development version

[tjnaughton]

Hi,

The new parser implementation in the development version seems to work perfectly for me, but I have noticed one regression compared to the latest released version of Brython.

>> a = 5
>> b = 6
>> print(f'{a=} {b=}')
a=5 a=} {b=6

I found it in the development version from 2024-02-10, but I've just verified it's still there as of 2024-02-25.

Thank you!