Merge BINARY_SUBSCR_LIST_INT with BINARY_SUBSCR_LIST_TUPLE

[eendebakpt]

The implementations of BINARY_SUBSCR_LIST_INT and BINARY_SUBSCR_TUPLE_INT are almost identical. They can be merged, so there is one opcode less. A branch with concept code is https://github.com/python/cpython/compare/main...eendebakpt:combine_list_tuple_opcode?expand=1

Is there value in reducing the number of opcodes in this way?

Microbenchmarks

import pyperf
runner = pyperf.Runner()

setup='l=[1,2,3,4,5,6,7,8,9,10]; t=(1,2,3,4)'

runner.timeit(name=f"list", stmt=f"x=l[3]",setup=setup)
runner.timeit(name=f"tuple", stmt=f"x=t[2]",setup=setup)
code="""
for ii in range(500):
	idx=ii%4
	a=l[idx]
	b=t[idx]	
"""
runner.timeit(name=f"mixed", stmt=code,setup=setup)

Results

list: Mean +- std dev: [base] 14.6 ns +- 0.5 ns -> [merged] 14.4 ns +- 0.1 ns: 1.01x faster
tuple: Mean +- std dev: [base] 14.6 ns +- 0.1 ns -> [merged] 14.4 ns +- 0.3 ns: 1.02x faster

Benchmark hidden because not significant (1): mixed

Geometric mean: 1.01x faster