Enhance profile setup: add cached/creating indicators, parallelize creation, and show online progress

[Copilot]

Profile setup progress was opaque and sequential. Users couldn't distinguish between cached profile reuse and fresh creation, multiple profiles were created serially, and there was no visibility into profiles coming online.

Changes

Progress indicators

• get_relay_profile() now returns (profile, is_cached) tuple
• Progress messages display: "Setting up profiles: X/Y (cached: A, creating: B)"
• Online progress shows: "Waiting for N profile(s) to come online: X/N" with real-time updates

Parallelization

• Receiver profile setup runs in concurrent threads
• Profile lookup/creation serialized under ProfileMaker.lock to prevent race conditions
• start_io() (the I/O-bound bottleneck) executes outside lock for true parallelization

Event handling

• ERROR events displayed during profile setup with detailed messages
• WARNING events logged during profile setup
• Progress bar continues updating as profiles come online

Terminology

• Renamed AccountMaker → ProfileMaker, get_relay_account → get_relay_profile
• Updated all user-facing strings and variables
• Preserved deltachat-rpc API: get_all_accounts(), add_account(), accounts_dir unchanged

Optimized thread safety

• Removed unnecessary lock from _add_online() (Python list.append() is atomic)
• Kept lock only in get_relay_profile() to prevent duplicate profile selection
• deltachat-rpc-server handles all API calls thread-safely

Implementation

class ProfileMaker:
    def __init__(self, dc):
        self.dc = dc
        self.online = []
        self.lock = threading.Lock()  # Only for profile selection
    
    def get_relay_profile(self, domain):
        with self.lock:
            # Find cached profile or create new one (prevent race)
            # ...
        # I/O runs outside lock - parallelizable
        self._add_online(profile)
        return profile, is_cached
    
    def wait_all_online(self):
        # Show progress: "Waiting for N profile(s) to come online: X/N"
        # Display ERROR/WARNING events as they occur

Threading strategy: serialize fast atomic operations (profile lookup/creation), parallelize slow I/O operations (start_io()). Minimal locking leverages Python's atomic operations and deltachat-rpc-server's thread safety.

Original prompt

during account setup make the progress spinner indicate better which accounts are already configured and cached, and which are freshly created.

parallelize the setup of fresh accounts
rename "account" to "profile" but take care to not change the API to deltachat-rpc objects.

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

[hpk42]

@copilot print progress bar for accounts coming online, and print the precise number instead of "all profiles". Show any error output that arrives via events. Read https://py.delta.chat for documentation of events.

[Copilot]

@copilot print progress bar for accounts coming online, and print the precise number instead of "all profiles". Show any error output that arrives via events. Read https://py.delta.chat for documenta...

Done in commit 6f789d0. The wait_all_online method now:

• Shows precise count: "Waiting for N profile(s) to come online: X/N"
• Updates progress in real-time as each profile comes online
• Displays ERROR events with detailed messages
• Displays WARNING events during setup

Example output:

# Setting up profiles: 3/3 (cached: 1, creating: 2) - Complete!
# Waiting for 3 profile(s) to come online: 0/3
# Waiting for 3 profile(s) to come online: 1/3
# Waiting for 3 profile(s) to come online: 2/3
# Waiting for 3 profile(s) to come online: 3/3

If errors occur during setup, they're displayed inline:

✗ ERROR during profile setup: <error message>
⚠ WARNING during profile setup: <warning message>