Track EVM stack "end" instead of "top"

[chfast]

This is more correct C++ because it avoids the "bottom" pointer (one below the stack space base pointer). In C++ only pointers to the elements of an array and the "end" pointers are well defined.

This also potentially helps with the "overaligned" stack space optimization. To be continued...

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 94.54%. Comparing base (53bcba6) to head (e9bb059).
Report is 1 commits behind head on master.

Additional details and impacted files

@@ Coverage Diff @@ ## master #895 +/- ## ======================================= Coverage 94.54% 94.54% ======================================= Files 175 175 Lines 19673 19674 +1 ======================================= + Hits 18599 18600 +1  Misses 1074 1074 

Flag	Coverage Δ
eest_gmp	15.28% <59.09%> (-0.01%)	⬇️
eof_execution_spec_tests	19.77% <83.33%> (-0.01%)	⬇️
ethereum_tests	21.34% <58.33%> (-0.01%)	⬇️
ethereum_tests_silkpre	18.29% <59.09%> (-0.01%)	⬇️
execution_spec_tests	18.44% <58.33%> (-0.01%)	⬇️
unittests	91.95% <100.00%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
lib/evmone/advanced_analysis.hpp	96.15% <100.00%> (ø)
lib/evmone/baseline_execution.cpp	100.00% <100.00%> (ø)
lib/evmone/execution_state.hpp	95.52% <100.00%> (-0.07%)	⬇️
lib/evmone/instructions.hpp	100.00% <100.00%> (ø)

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[chfast]

Seems no performance regressions:

Zen2, clang-18:

baseline/execute/main/blake2b_huff/empty_mean +0.0013 +0.0013 10 10 10 10 baseline/execute/main/blake2b_huff/8415nulls_mean -0.0105 -0.0105 603 597 603 597 baseline/execute/main/blake2b_shifts/8415nulls_mean -0.0268 -0.0268 5650 5498 5650 5498 baseline/execute/main/sha1_divs/empty_mean -0.0080 -0.0080 42 42 42 42 baseline/execute/main/sha1_divs/5311_mean +0.0010 +0.0010 3335 3338 3335 3338 baseline/execute/main/sha1_shifts/empty_mean -0.0070 -0.0070 22 21 22 21 baseline/execute/main/sha1_shifts/5311_mean -0.0081 -0.0081 1776 1761 1776 1761 baseline/execute/main/snailtracer/benchmark_mean -0.0088 -0.0088 29799 29538 29799 29538 baseline/execute/main/structarray_alloc/nfts_rank_mean -0.0136 -0.0136 396 391 396 391 baseline/execute/main/swap_math/spent_mean -0.0081 -0.0081 2 2 2 2 baseline/execute/main/swap_math/received_mean -0.0133 -0.0133 2 2 2 2 baseline/execute/main/swap_math/insufficient_liquidity_mean -0.0034 -0.0034 1 1 1 1 baseline/execute/main/weierstrudel/1_mean -0.0019 -0.0019 139 139 139 139 baseline/execute/main/weierstrudel/15_mean -0.0119 -0.0119 1387 1370 1387 1370 OVERALL_GEOMEAN -0.0085 -0.0085 0 0 0 0 

[pdobacz]

codewise LGTM, but what does this accomplish? (put in the pr description too)