Further optimize SlottedArray

[benaadams]

Doesn't need to be two object references + length; can be 1 object ref + data length

Benchmarks

Before

Method	Mean	Error	StdDev	Code Size	Allocated
Write_whole_page_of_data	10.067 us	0.0995 us	0.0930 us	2,458 B	-
Read_existing_keys_prefix_different	11.866 us	0.1176 us	0.1100 us	1,425 B	-
Read_existing_keys_suffix_different	9.123 us	0.0830 us	0.0735 us	1,439 B	-
Read_nonexistent_keys	13.898 us	0.0493 us	0.0437 us	1,431 B	-
Hash_collisions	18.141 us	0.3537 us	0.6287 us	3,329 B	-

After

Method	Mean	Error	StdDev	Code Size	Allocated
Write_whole_page_of_data	9.890 us	0.1350 us	0.1263 us	2,525 B	-
Read_existing_keys_prefix_different	10.573 us	0.1386 us	0.1296 us	1,556 B	-
Read_existing_keys_suffix_different	9.196 us	0.1169 us	0.1093 us	1,471 B	-
Read_nonexistent_keys	13.716 us	0.1773 us	0.1659 us	1,480 B	-
Hash_collisions	51.047 us	0.6043 us	0.5653 us	3,439 B	-

[Scooletz]

@benaadams Looks like not that much gains and the Hash_collisions benchmark went through the roof? Would you have some time to take a look at it? I've added results to the description

[github-actions]

Package	Line Rate	Branch Rate	Health
Paprika	85%	79%	➖
Summary	85% (3559 / 4191)	79% (1102 / 1394)	➖

Minimum allowed line rate is 75%