[PERF]: Epic for binding overhead improvements

This issue is tracking performance improvements and investigations to Python-to-C binding overhead, mostly driven by the benchmark of cuTensorMapEncodeTiled devised in #659. That is a useful benchmark because it is a function with an unusually high number of arguments (and therefore unusually high Python-to-C overhead).

Comparison to a more limited Cython binding

As an interesting experimental datapoint, a colleague provided a vibe-coded Cython binding for cuTensorMapEncodeTiled that runs about 4x faster than cuda-bindings official one. It is useful to see where some overheads may be reduced, but care should be taken looking at its raw performance: this wrapper accepts far fewer things as inputs than the CUDA bindings, and doesn't include developer niceties, like enums.

Merged or in-progress fixes

Timings below are per-iteration of the benchmark in #659. This includes /both/ binding overhead and some fixed amount of time in the actual CUDA call.

• 4.80us Baseline time
• 3.63us #1543
• 2.73us #1545
• 2.70us #1581
• 2.59us #1616
• 2.38us #1638
• (no change on this benchmark) #1644

Under investigation

Issues in this category are theoretical findings to reduce the operations required for type conversion, but haven't necessarily yet been confirmed to have a measurable effect.

• #1639
• #1640
• #1642

Deferred (effective, but high effort)

• #1643

Rejected (ineffective)

• #1605
• #1649
• #1637