Revert "Partial Fix Issue 16615 - convert os pid to std.process Pid"

[CyberShadow]

Reverts #5086

See #5086 (comment) for rationale.

[dlang-bot]

Fix	Bugzilla	Description
✓	16615	std.process is missing functionality for child processes

[wilzbach]

I am just following the discussion passively, but it's clear that that it's a very active open discusison, so reverting it & continuing the discussion sounds very reasonable.

[andralex]

I would appreciate if my decisions weren't undone without my approval. Thanks.

[wilzbach]

I would appreciate if my decisions weren't undone without my approval. Thanks.

Sorry, I thought it's the general policy that when a controversial PR was merged without appropriate review and/or someone pointing out non-trivial issues that we revert the change until there's a consensus on the best course of action?

[CyberShadow]

I would appreciate if my decisions weren't undone without my approval. Thanks.

Considering that reviewer time is scarce, I think it's understandable when occasionally something is merged with a superficial review by someone without experience in said area of the code base. However, the consequence of that is that we should be OK with reverting PRs as soon as it becomes obvious that there are more issues that need additional scrutiny post merge, which granted should've been raised during review in a perfect world. I think reverting early is a good idea, as it avoids the risk of an unfinished PR making it into the stable branch, in which case it takes more work of getting it out.