gh-150932: Add a partial-line-marker on Apple system logger and optimize the iOS testbed runner

[timrid]

This adds a marker for partial lines on the Apple system logger, so that a cooperating log reader (like the iOS testbed runner) can re-join them instead of rendering each flush as a separate line in the Apple system log.

Additionally, this optimizes the iOS testbed runner by:

• Joining partial lines marked by the system log, so that they are rendered as a single line in the testbed output.
• Restoring control characters (e.g. ANSI color escapes) escaped by the system log.

Fixes #150932

---

To test this, I used this test script and ran it in an iOS testbed:

# Test colors ###################################
colors = {
    "red":     "\033[31m",
    "green":   "\033[32m",
    "yellow":  "\033[33m",
    "blue":    "\033[34m",
    "magenta": "\033[35m",
    "cyan":    "\033[36m",
    "white":   "\033[37m",
}
reset = "\033[0m"

for name, code in colors.items():
    print(f"{code}Hello in {name}!{reset}")

# Test newline handling #########################
print("This should be ", end="", flush=True)
print("in one line.")

print("This should al", end="", flush=True)
print("so be in one", end="", flush=True)
print(" line.")

print("This are \n two lines.")

# Test live output ##############################
import time
for i in range(10):
    print(".", end="")
    time.sleep(1)

• Issue: Strange new line behavior and missing color in the iOS testbed logs #150932

[bedevere-app]

Most changes to Python require a NEWS entry. Add one using the blurb_it web app or the blurb command-line tool.

If this change has little impact on Python users, wait for a maintainer to apply the skip news label instead.

[freakboy3742]

Thanks for the PR; however, I'm going to close this on the basis that the discussion on the underlying ticket hasn't yet concluded that either of these ideas are desirable.

Of the two, the ANSI reversal is a firm "no" from me. I'm not yet convinced the line-end thing is worthwhile either; but if I am convinced, it will need to be extracted from the ANSI reversal stuff, so a clean PR would make sense.