Fix developer-cli run command leaving worker processes alive on Linux/macOS

[tjementum]

Summary & Motivation

Fix run --stop failing to terminate worker processes (Account.Workers, BackOffice.Workers, Main.Workers) on Linux/macOS. After stopping, these processes remained listening on their ports indefinitely.

The previous stop logic used a two-loop approach that had two compounding bugs:

• The first loop only scanned ports at multiples of 100 (9100, 9200, 9300) to discover process names, but workers listen on different ports (e.g., 9199, 9299, 9399) — so the known-names set was always empty when workers were the only survivors
• The second loop used ps -o comm= which truncates binary names on Linux (e.g., BackOffice.Work instead of BackOffice.Workers), causing substring checks like "PlatformP" to miss them
• The pkill -9 -f aspire and pkill -9 -f dcp fallback calls killed ALL Aspire/dcp processes system-wide, which is unsafe when running multiple Aspire projects

Replace the entire approach with a single loop that uses ps -o args= (full command line) and matches against Configuration.SourceCodeFolder. This reliably catches all application processes regardless of binary name or port, while being safe for multi-project setups since each project has a unique source folder path.

Checklist

• I have added tests, or done manual regression tests
• I have updated the documentation, if necessary

[linear]

PP-1076 Fix developer cli leaving worker processes alive on Linux/macOS