Optimize shadow stack instruction sequences

[Robbepop]

Closes #920.

TODOs

• Fuse i32.add_imm with global.set 0
• Fuse global.get 0 with i32.add_imm
• Fuse fused global.get 0 + i32.add_imm with i32.local_tee and global.set 0
• Support fusion of i32.add with a function local constant rhs register.

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 77.57848% with 50 lines in your changes are missing coverage. Please review.

Project coverage is 81.79%. Comparing base (f626385) to head (15cd321).

Files	Patch %	Lines
...rates/wasmi/src/engine/translator/instr_encoder.rs	72.88%	16 Missing ⚠️
crates/wasmi/src/engine/executor/instrs/global.rs	30.00%	14 Missing ⚠️
crates/wasmi/src/engine/translator/visit.rs	55.00%	9 Missing ⚠️
crates/wasmi/src/engine/executor/instrs.rs	33.33%	4 Missing ⚠️
...ates/wasmi/src/engine/translator/visit_register.rs	0.00%	3 Missing ⚠️
crates/wasmi/src/engine/translator/mod.rs	90.00%	2 Missing ⚠️
...rates/wasmi/src/engine/translator/relink_result.rs	33.33%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #928      +/-   ##
==========================================
- Coverage   81.79%   81.79%   -0.01%     
==========================================
  Files         262      262              
  Lines       24302    24499     +197     
==========================================
+ Hits        19878    20039     +161     
- Misses       4424     4460      +36     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[Robbepop]

Comment on why this has not yet been merged despite passing all CI tests and working: it introduces a lot of complexity into the translation pipelined compared to the gains we see and can verify. The runtime gains are mostly single digits and restricted to sets of benchmarks that actually make use of the shadow stack. The global_bump benchmark is heavily affected with roughly 30% performance improvement but it is also very artificial. On the flip side the improvements to memory consumption also exist but are also just single digits for practical Wasm binaries.

So all in all the question is whether the gains are worth the added complexities.

[Robbepop]

The miri CI job fails because of this issue: rust-lang/miri#3404