Renew performance probes (July 2021)

[mcomella]

The following probes are up for expiration:

start_reason_process_error
start_reason_activity_error

Startup.timeline:
  framework_primary
  framework_secondary
  framework_start_error
  framework_start_read_error
  clock_ticks_per_second_v2

perf.startup:
 cold_main_app_to_first_frame:
 cold_view_app_to_first_frame
 cold_unknwn_app_to_first_frame
 application_on_create
 app_on_create_to_glean_init
 app_on_create_to_megazord_init
 app_on_create_to_setup_in_main
 home_activity_on_create
 home_activity_on_start
 home_fragment_on_create_view
 home_fragment_on_view_created
 base_bfragment_on_create_view
 base_bfragment_on_view_created

storage.stats:
 query_stats_duration
 app_bytes
 cache_bytes
 data_dir_bytes

We should renew them fast before we miss a release if we're still interested in them. Though we may want to transfer ownership of storage.stats to the fenix team since we don't actively track it – the main problem we have there is that we don't know how much is too much disk space.

┆Issue is synchronized with this Jira Task

[mcomella]

eliserichards put up a PR that should make this easier: #20519

[MarcLeclair]

We've established that we would keep the following probes from perf.startup:

cold_main_app_to_first_frame:
 cold_view_app_to_first_frame
 cold_unknwn_app_to_first_frame
 application_on_create

We also decided to keep all of storage.stats and remove Startup.timeline

[mcomella]

We'll also keep:

start_reason_process_error
start_reason_activity_error

As they monitor errors for the cold_*_to_first_frame probes.

[mcomella]

I filed #20596 for the probe removal.

[mcomella]

I created a commit with our needed renewal mcomella@d670e30 based on eliserichards' PR: #20519 I'm discussing with eliserichards about who should take care of merging this.

[mcomella]

Dupe #20518 (because all of the commits in my PR point at that issue even though this one has more info).