Fix snuba-api healthcheck bypassing proxy for local requests

[oddballfr]

Fix snuba-api healthcheck bypassing proxy for local requests (urllib NO_PROXY doesn't support CIDR ranges)

Problem

The snuba-api healthcheck (api_healthcheck.py) was failing in environments with an HTTP(S) proxy configured (HTTP_PROXY/HTTPS_PROXY). Squid logs confirmed the local request was being routed through the proxy and rejected:
TCP_DENIED/403 ... GET http://127.0.0.1:1218/health

Root cause

urllib.request does not support CIDR notation (127.0.0.0/8) in NO_PROXY — it only matches exact hostnames or domain suffixes. A NO_PROXY value that works for curl/wget can therefore still leave this local request routed through the configured proxy.

Fix

self-hosted/snuba/api_healthcheck.py now uses a dedicated urllib.request.build_opener(urllib.request.ProxyHandler({})) opener, explicitly forcing no proxy for this local request instead of relying on NO_PROXY parsing.

Testing

snuba-api healthcheck passes again after container rebuild, in an environment with HTTP_PROXY/HTTPS_PROXY configured.

Legal Boilerplate

Look, I get it. The entity doing business as "Sentry" was incorporated in the State of Delaware in 2015 as Functional Software, Inc. and is gonna need some rights from me in order to utilize my contributions in this here PR. So here's the deal: I retain all rights, title and interest in and to my contributions, and by keeping this boilerplate intact I confirm that Sentry can use, modify, copy, and redistribute my contributions, under Sentry's choice of terms.

[aminvakil]

I think it's better to set NO_PROXY to 127.0.0.1 and there is no need for this change.

This should be documented in container scope proxy setting in docs though, current docs only say 127.0.0.0/8.

[oddballfr]

Thanks for the review.
I actually tried adding 127.0.0.1 to NO_PROXY but the healthcheck kept failing.
This fix is what unblocked me, both on v25.5.2 (when the Python healthcheck was introduced) and again on v26.7.0.