Refactor encoding check as per brotli module

[robvanoostenrijk]

Using zstd-nginx-module with the following configuration, does not work on for example autoindex generated documents.

        zstd			on;
        zstd_min_length		64;
        zstd_comp_level		8;
	zstd_types		application/javascript
				application/json
				application/xhtml+xml
				application/xml
				font/opentype
				font/ttf
				image/svg+xml
				image/x-icon
				text/css
				text/plain;

	location /folder {
		alias			/mnt/web/folder;
		autoindex		on;
		autoindex_format	xml;

		xslt_string_param	footer $host;
		xslt_string_param	path $uri;
		xslt_string_param	layout $arg_layout;
		xslt_string_param	sort $arg_sort;
		xslt_string_param	order $arg_order;

		xslt_stylesheet		/etc/nginx/file_index.xslt;
	}

This request returns the following on request:

> GET /folder/ HTTP/2
> Host: server.example
> User-Agent: curl/8.11.0
> Accept: */*
> Accept-encoding: zstd
>
< HTTP/2 200
< date: Sun, 22 Dec 2024 14:54:58 GMT
< content-type: text/html; charset=UTF-8
< content-length: 43743
....

Updating the logic where the filter determines if its going to compress the response, aligning to what Google ngx_brotli uses, resolves the above issue.

Additionally lifting out the request check function to a shared code file ensures the logic is identical for both static and dynamic filters.

[SuperSandro2000]

This request returns the following on request:

Can you point the issue with the response? I cannot pin point it.

[robvanoostenrijk]

This request returns the following on request:

Can you point the issue with the response? I cannot pin point it.

Even though the request sets Accept-Encoding: zstd correctly, the response is not Content-Encoding: zstd.

This is mostly an issue with dynamically generated nginx content, static pages and/or files work fine in the original code.
Likely the charset=UTF-8 throws the parsing off, hence borrowing the brotli parsing as it does not have this issue.