feat: modularize toolchain tests

[arayikhalatyan]

Resolves INT-2740

Moved the tests in ./crates/toolchain/tests into respective test crates inside the corresponding extensions

[jonathanpwang]

We should update the layout.md doc too

[jonathanpwang]

Overall refactor and modularization looks good. I had some inline comments but also we should:

1. Add the cargo nextest to each of the respective github workflows (e.g., ecc.yml), right now the tests won't actually be run.
2. Can we add for each of the examples linked from the book, the correct openvm.toml to the test programs dir?
3. After (2), test locally that the cli works, then let's add tests to cli.yml where we run the cli from the test guest program dir exactly as one should do it from the book. This way CI will check that the book is up to date without us having to manually check.

[jonathanpwang]

delete

[jonathanpwang]

LGTM except some minor CI cleanup

[github-actions]

Benchmarks

group	app_log_blowup	app_total_cells_used	app_total_cycles	app_total_proof_time_ms	leaf_log_blowup	leaf_total_cells_used	leaf_total_cycles	leaf_total_proof_time_ms	max_segment_length	instance	alloc
ecrecover_program	2	10,251,804	195,066	(+152.0 [+8.1%]) 2,029.0	-	-	-	-	1048476	64cpu-linux-arm64	mimalloc
fibonacci_program	2	51,615,800	3,000,274	(-9.0 [-0.2%]) 5,519.0	-	-	-	-	1048476	64cpu-linux-arm64	mimalloc
regex_program	2	238,890,449	8,381,808	(-341.0 [-1.9%]) 17,147.0	-	-	-	-	1048476	64cpu-linux-arm64	mimalloc
verify_fibair	2	48,127,147	397,164	(-22.0 [-0.7%]) 3,153.0	-	-	-	-	1048476	64cpu-linux-arm64	mimalloc

Commit: 00ba22f

Benchmark Workflow