Looking at the code as a whole, we ended up with a little to much "buttons". It turns out that doing case insensitive matching (lower both pattern and strings to match) performs just fine. Or at least, it
gives the penalty to the people who uses mixed case filenames.
```
GetGlob/Default_cache-10 10.6ns ± 2% 10.6ns ± 1% ~ (p=0.657 n=4+4)
GetGlob/Filenames_cache,_lowercase_searchs-10 10.6ns ± 2% 10.6ns ± 0% ~ (p=1.000 n=4+4)
GetGlob/Filenames_cache,_mixed_case_searchs-10 29.7ns ± 1% 29.6ns ± 1% ~ (p=0.886 n=4+4)
GetGlob/GetGlob-10 13.7ns ± 1% 13.7ns ± 0% ~ (p=0.429 n=4+4)
name old alloc/op new alloc/op delta
GetGlob/Default_cache-10 0.00B 0.00B ~ (all equal)
GetGlob/Filenames_cache,_lowercase_searchs-10 0.00B 0.00B ~ (all equal)
GetGlob/Filenames_cache,_mixed_case_searchs-10 5.00B ± 0% 5.00B ± 0% ~ (all equal)
GetGlob/GetGlob-10 0.00B 0.00B ~ (all equal)
name old allocs/op new allocs/op delta
GetGlob/Default_cache-10 0.00 0.00 ~ (all equal)
GetGlob/Filenames_cache,_lowercase_searchs-10 0.00 0.00 ~ (all equal)
GetGlob/Filenames_cache,_mixed_case_searchs-10 1.00 ± 0% 1.00 ± 0% ~ (all equal)
GetGlob/GetGlob-10
```
The old implementation had some issues, mostly related to the context (e.g. name, file paths) passed to the template.
This new implementation is using the exact same code path for evaluating the pages as in a regular build.
This also makes it more robust and easier to reason about in a multilingual setup.
Now, if you are explicit about the target path, Hugo will now always pick the correct mount and language:
```bash
hugo new content/en/posts/my-first-post.md
```
Fixes#9032Fixes#7589Fixes#9043Fixes#9046Fixes#9047
