[BUG] Timeouts do not work with TLS or HTTPS connections

[lukaspustina]

Describe the bug
When using DNS servers with TLS or HTTPS connection types the configured timeout are not applied.

To Reproduce
Steps to reproduce the behavior:

• mhost --no-system-lookups -s tls:1.1.1.1:852,tls_auth_name=dns.quad9.net,name="Quad9 1" lookup example.com
• mhost --no-system-lookups -s https:1.1.1.1:852,tls_auth_name=dns.quad9.net,name="Quad9 1" lookup example.com

The ports are wrong in the above examples and mhost should abort due to the default timeout after 5 seconds. but it takes over a minute.

[15:04:29 lukas] ~  →  mhost --no-system-lookups -s tls:1.1.1.1:852,tls_auth_name=dns.quad9.net,name="Quad9 1" lookup example.com
▶︎ Options: mode=multi, concurrent nameservers=10, max. nameservers=100, concurrent requests=5, retries=0, timeout=5s, ndots=1, abort on error, abort on timeout
▶︎ Running lookups.
▸ Sending 4 requests to 1 nameserver for 4 record types of 1 name.
▸ Received 0 responses with 0 RR [], 0 Nx, 1 Err [1 TO, 0 QR, 0 SF, 0 O] in (min -, max -) ms from 0 servers within 76204 ms of total run time.
❖ Finished.
[15:05:49 lukas] ~  →

Expected behavior
mhost should abort after the timeout which in these examples is the default timeout of 5 seconds.

Version (please complete the following information):

• OS: Both, Linux and macos
• mhost 0.3.0-alpha-1

[lukaspustina]

This is a known issue in trust-dns resolver; cf. https://github.com/bluejekyll/trust-dns/issues/1073