Delete special handling of per tensor quant in `_choose_scale_float8`

[jerryzh168]

Recently I found that for _choose_scale_float8:

len(block_size) == 0 for per tensor:

ao/torchao/quantization/quant_primitives.py

Line 2303 in b4ec4cb

if len(block_size) == 0:

This seems unnecessary, and I feel we should merge this branch with the others

[KaranPradhan266]

@jerryzh168 I’d love to take this on, may I work on this issue?