Currently ExecutionEngine tries to dump all functions declared in the
module, even those which are "external" (i.e., linked/loaded at
runtime). E.g.

```mlir
func.func private @printF32(f32)
func.func @supported_arg_types(%arg0: i32, %arg1: f32) {
  call @printF32(%arg1) : (f32) -> ()
  return
}
```
fails with
```
Could not compile printF32:
  Symbols not found: [ __mlir_printF32 ]
Program aborted due to an unhandled Error:
Symbols not found: [ __mlir_printF32 ]
```
even though `printF32` can be provided at final build time (i.e., when
the object file is linked to some executable or shlib). E.g, if our own
`libmlir_c_runner_utils` is linked.

So just skip functions which have no bodies during dump (i.e., are decls
without defns).

In order to optionally run some checks that depend on the `ml_dtypes`
python module we have to remove the `CHECK` lines for those tests or
they will be required and missed in the test output.

I've changed to use asserts as recommended in [1].

[1]:
https://github.com/llvm/llvm-project/pull/123061#issuecomment-2596116023

We noticed that `mlir/python/requirements.txt` lists `ml_dtypes` as a requirement but when looking at the code in `mlir/python`, the only `import` is guarded:

```python
try:
    import ml_dtypes
except ModuleNotFoundError:
    # The third-party ml_dtypes provides some optional low precision data-types for NumPy.
    ml_dtypes = None
```

This makes `ml_dtypes` an optional dependency.

Some python tests however partially depend on `ml_dtypes` and should not run if that module is unavailable. That is what this change does.

This is a replacement for #123051 which was excluding tests too broadly.

Description:
- removed trailing spaces in few files
- fixed markdown link definition:

Do not run `cf-to-llvm` as part of `func-to-llvm`. This commit fixes
https://github.com/llvm/llvm-project/issues/70982.

This commit changes the way how `func.func` ops are lowered to LLVM.
Previously, the signature of the entire region (i.e., entry block and
all other blocks in the `func.func` op) was converted as part of the
`func.func` lowering pattern.

Now, only the entry block is converted. The remaining block signatures
are converted together with `cf.br` and `cf.cond_br` as part of
`cf-to-llvm`. All unstructured control flow is not converted as part of
a single pass (`cf-to-llvm`). `func-to-llvm` no longer deals with
unstructured control flow.

Also add more test cases for control flow dialect ops.

Note: This PR is in preparation of #120431, which adds an additional
GPU-specific lowering for `cf.assert`. This was a problem because
`cf.assert` used to be converted as part of `func-to-llvm`.

Note for LLVM integration: If you see failures, add
`-convert-cf-to-llvm` to your pass pipeline.

Do not run `arith-to-llvm` as part of `func-to-llvm`. This commit partly
fixes #70982.

Also simplify the pass pipeline for two math dialect integration tests.

Note for LLVM integration: If you see failures, add `arith-to-llvm` to your pass pipeline.

Prefer to get the path to libmlir_runner_utils and
libmlir_c_runner_utils via %mlir_runner_utils and %mlir_c_runner_utils.
Fallback to the previous paths only if they aren't defined.

This ensures the test will pass regardless of the build configuration
used downstream.

add f8E5M2 and tests for np_to_memref

---------

Co-authored-by: Zhicheng Xiong <zhichengx@dc2-sim-c01-215.nvidia.com>

Reapply the PR which was reverted due to built-bots, and now the bots
get updated.
https://discourse.llvm.org/t/need-a-help-with-the-built-bots/79437
original PR: https://github.com/llvm/llvm-project/pull/92489, reverted
in https://github.com/llvm/llvm-project/pull/93771

Reverts llvm/llvm-project#92489

This broke the bots.