Rollup merge of #136083 - bend-n:⃤⃤, r=lcnr

Suggest {to,from}_ne_bytes for transmutations between arrays and integers, etc

implements #136067

Rust has helper methods for many kinds of safe transmutes, for example integer<->bytes. This is a lint against using transmute for these cases.

```rs
fn bytes_at_home(x: [u8; 4]) -> u32 {
   transmute(x)
}

// other examples
transmute::<[u8; 2], u16>();
transmute::<[u8; 8], f64>();
transmute::<u32, [u8; 4]>();
transmute::<char, u32>();
transmute::<u32, char>();
```
It would be handy to suggest `u32::from_ne_bytes(x)`.
This is implemented for `[u8; _]` -> `{float int}`

This also implements the cases:
`fXX` <-> `uXX` = `{from_bits, to_bits}`
`uXX` -> `iXX` via `cast_unsigned` and `cast_signed`
{`char` -> `u32`, `bool` -> `n8`} via `from`
`u32` -> `char` via `from_u32_unchecked` (note: notes `from_u32().unwrap()`) (contested)
`u8` -> `bool` via `==` (debatable)

---
try-job: aarch64-gnu
try-job: test-various

Author: matthiaskrgr

tests/fail/validity/invalid_bool.rs

@@ -1,3 +1,4 @@
+#![allow(unnecessary_transmutes)]
 fn main() {
     let _b = unsafe { std::mem::transmute::<u8, bool>(2) }; //~ ERROR: expected a boolean
 }

tests/fail/validity/invalid_bool_op.rs

@@ -2,6 +2,7 @@
 // Make sure we find these even with many checks disabled.
 //@compile-flags: -Zmiri-disable-alignment-check -Zmiri-disable-stacked-borrows -Zmiri-disable-validation
 
+#![allow(unnecessary_transmutes)]
 fn main() {
     let b = unsafe { std::mem::transmute::<u8, bool>(2) };
     let _x = b == std::hint::black_box(true); //~ ERROR: interpreting an invalid 8-bit value as a bool

tests/fail/validity/invalid_char.rs

@@ -1,3 +1,4 @@
+#![allow(unnecessary_transmutes)]
 fn main() {
     assert!(std::char::from_u32(-1_i32 as u32).is_none());
     let _val = match unsafe { std::mem::transmute::<i32, char>(-1) } {

tests/fail/validity/invalid_char_op.rs

@@ -2,6 +2,7 @@
 // Make sure we find these even with many checks disabled.
 //@compile-flags: -Zmiri-disable-alignment-check -Zmiri-disable-stacked-borrows -Zmiri-disable-validation
 
+#![allow(unnecessary_transmutes)]
 fn main() {
     let c = 0xFFFFFFu32;
     assert!(std::char::from_u32(c).is_none());

tests/pass/float.rs

@@ -6,6 +6,7 @@
 #![feature(f16)]
 #![allow(arithmetic_overflow)]
 #![allow(internal_features)]
+#![allow(unnecessary_transmutes)]
 
 use std::any::type_name;
 use std::cmp::min;

tests/pass/issues/issue-miri-184.rs

@@ -1,3 +1,4 @@
+#![allow(unnecessary_transmutes)]
 pub fn main() {
     let bytes: [u8; 8] = unsafe { ::std::mem::transmute(0u64) };
     let _val: &[u8] = &bytes;

tests/pass/shims/x86/intrinsics-x86-gfni.rs

@@ -368,7 +368,7 @@ unsafe fn load_m256i_word<T>(data: &[T], word_index: usize) -> __m256i {
 #[target_feature(enable = "avx512f")]
 unsafe fn load_m512i_word<T>(data: &[T], word_index: usize) -> __m512i {
     let byte_offset = word_index * 64 / size_of::<T>();
-    let pointer = data.as_ptr().add(byte_offset) as *const i32;
+    let pointer = data.as_ptr().add(byte_offset) as *const __m512i;
     _mm512_loadu_si512(black_box(pointer))
 }
 

tests/pass/shims/x86/intrinsics-x86-sse.rs

@@ -1,5 +1,6 @@
 // We're testing x86 target specific features
 //@only-target: x86_64 i686
+#![allow(unnecessary_transmutes)]
 
 #[cfg(target_arch = "x86")]
 use std::arch::x86::*;