Implement simple benchmarks for instantiation.

[peterhuene]

This adds benchmarks around module instantiation using criterion.

Both the default (i.e. on-demand) and pooling allocators are tested
sequentially and in parallel using a thread pool.

Instantiation is tested with an empty module, a module with a single page
linear memory, a larger linear memory with a data initializer, and a "hello
world" Rust WASI program.

[peterhuene]

Note: as the pooling/uffd allocator requires multiple instantiations from the same engine instance, these benchmarks were not implemented with sightglass yet.

[abrown]

(I'm just waiting to see if cargo-deny blows up like it did for me over in #2888...)

[peterhuene]

It did indeed :) There's no rush on merging this, so I'll do so after your PR goes in.

[alexcrichton]

FWIW from CI:

sequential/default/empty.wat
                        time:   [3.0138 us 3.0327 us 3.0524 us]
sequential/default/small_memory.wat
                        time:   [9.7558 us 9.9036 us 10.053 us]
sequential/default/data_segments.wat
                        time:   [20.643 us 21.009 us 21.425 us]
sequential/default/wasi.wasm
                        time:   [29.953 us 30.328 us 30.756 us]
sequential/pooling/empty.wat
                        time:   [3.1582 us 3.1839 us 3.2118 us]
sequential/pooling/small_memory.wat
                        time:   [5.2716 us 5.3406 us 5.4295 us]
sequential/pooling/data_segments.wat
                        time:   [14.172 us 14.299 us 14.444 us]
sequential/pooling/wasi.wasm
                        time:   [26.612 us 26.868 us 27.165 us]
parallel/default/1000 instances with 1 thread
                        time:   [33.318 ms 33.611 ms 33.904 ms]
parallel/default/1000 instances with 2 threads
                        time:   [21.400 ms 21.664 ms 21.949 ms]
parallel/pooling/1000 instances with 1 thread
                        time:   [29.697 ms 30.111 ms 30.531 ms]
parallel/pooling/1000 instances with 2 threads
                        time:   [23.463 ms 23.723 ms 24.012 ms]
sequential/uffd/empty.wat
                        time:   [2.8096 us 2.8404 us 2.8799 us]
sequential/uffd/small_memory.wat
                        time:   [3.5588 us 3.5888 us 3.6222 us]
sequential/uffd/data_segments.wat
                        time:   [3.7279 us 3.7799 us 3.8324 us]
sequential/uffd/wasi.wasm
                        time:   [11.984 us 12.179 us 12.396 us]
parallel/uffd/1000 instances with 1 thread
                        time:   [14.558 ms 14.702 ms 14.848 ms]
parallel/uffd/1000 instances with 2 threads
                        time:   [11.417 ms 11.549 ms 11.694 ms]

[abrown]

#2888 is merged so perhaps a rebase would make the CI go green (I kicked off the CI in case that works...).