fuzzgen: Add SIMD instructions supported by the interpreter (#5971)

* fuzzgen: Add some SIMD instructions

* fuzzgen: Remove `scalar_to_vector`

Broken in the interpreter #5911

* fuzzgen: Remove SIMD bitcasts

Broken in the interpreter #5915

* fuzzgen: Fix insert lane

* fuzzgen: Remove debug code

* fuzzgen: Remove vall_true

This is broken in the interpreter #5916

* fuzzgen: Disable a few more ops

* fuzzgen: Remove `iadd_pairwise.i64x2`

Turns out it doesen't exist

* fuzzgen: Remove scalar `sqmul_round_sat`

#5923

* fuzzgen: Disable aligned loads to SIMD values

* fuzzgen: Address Review Feedback

Co-Authored-By: Jamey Sharp <jsharp@fastly.com>

* fuzzgen: Rework `cmp` exclusion rules

Co-Authored-By: Jamey Sharp <jsharp@fastly.com>

---------

Co-authored-by: Jamey Sharp <jsharp@fastly.com>