Improve performance by fusing kernels

[au2001]

Change Summary

What does this PR change?

It improves performance by 15-98% through 2 compounding changes:

• aligning mp_number to 2x 16 bytes instead of 8x 4 bytes; and
• merging kernels for iterating/transforming/scoring into a single one

(profanity_inverse is untouched in this PR, so there are still 2 launches total, down from 3-4)

Testing & Verification

How was this tested?

• Unit tests
• Integration tests
• Manual testing (describe steps)
• Verified on staging

I ran each scoring method on different GPUs with the previous and new codebase.
For each resulting address, I ran an external verification script which found no mistake.
Here are the median results for the ~most popular variant of each NVIDIA generation:

GPU	VRAM	base	head	inverse-multiple	gain
L40S	45 GB	1,228 MH/s	1,413 MH/s	-I 1572864	+15.1%
RTX 5070	12 GB	797 MH/s	1,397 MH/s	-I 393216	+75.2%
RTX 4060	8 GB	397 MH/s	623 MH/s	-I 229376	+56.7%
RTX 3060	12 GB	342 MH/s	676 MH/s	-I 458752	+97.7%
RTX 2060	12 GB	263 MH/s	476 MH/s	-I 8192	+81.1%
GTX 1060	3 GB	141 MH/s	263 MH/s	-I 98304	+86.2%

Notes:
GPUs which were bound by memory latency/bandwidth see the most performance increase.
That's consumer-grade GPUs, while the L40S sees the least gain, as expected.

During benchmarks, when reaching speeds of 2-3ms per round, the millisecond clock precision was insufficient.
I thus bumped it to microseconds for increased accuracy.

A bug also prevented from enabling profiling: clCreateCommandQueueWithProperties on OpenCL 2.0+ expects a zero terminated list of name/value pairs, not a bitfield like 1.2 – 0 just so happened to be interpreted as an empty list.

Risk Assessment

Risk Level:

• Low - Minor changes, no operational impact
• Medium - Moderate changes, limited impact, standard rollback available
• High - Significant changes, potential operational impact, complex rollback

Risks & Impact

Fully backwards-compatible with no user-facing changes.
But changes how the kernels work in a non-trivial way.
Requires clearing the cache to see changes (and to rollback).

Disclaimer: this PR was assisted by Claude Opus 5 for code exploration & benchmark scripts.
However, all code being submitted has been hand-written – and I'm a human :)