Clear all the local metrics data when user opt out of CEIP

[prkalle]

What this PR does / why we need it

This PR add changes to clear all the local metrics data when user opt out of CEIP and CLI will ignore (not store) the metrics.

Which issue(s) this PR fixes

Fixes #

Describe testing done for PR

Initially opted-in for the telemetry and later opted-out and verified the local metrics DB data is cleared.

❯ sqlite3 -batch ~/.config/tanzu-cli-telemetry/cli_metrics.db "select count(*) from tanzu_cli_operations;"
count(*)
0
❯ ./bin/tanzu telemetry cli-usage-analytics update --opt-in
I0118 16:12:52.175949   92176 update.go:70] "successfully opted into CEIP"
I0118 16:12:52.189164   92176 update.go:95] "successfully set CLI telemetry options"
❯ ./bin/tanzu context list --current
  NAME                ISACTIVE  TYPE   PROJECT  SPACE
  prem-ucp-intg-ctx7  true      tanzu

[i] Use '--wide' flag to view additional columns.
❯ ./bin/tanzu context list --current
  NAME                ISACTIVE  TYPE   PROJECT  SPACE
  prem-ucp-intg-ctx7  true      tanzu

[i] Use '--wide' flag to view additional columns.
❯ ./bin/tanzu context list --current
  NAME                ISACTIVE  TYPE   PROJECT  SPACE
  prem-ucp-intg-ctx7  true      tanzu

[i] Use '--wide' flag to view additional columns.
❯ sqlite3 -batch ~/.config/tanzu-cli-telemetry/cli_metrics.db "select count(*) from tanzu_cli_operations;"
count(*)
3
❯ ./bin/tanzu telemetry cli-usage-analytics update --opt-out
I0118 16:13:31.126997   92246 update.go:77] "successfully opted out of CEIP"
I0118 16:13:31.138951   92246 update.go:95] "successfully set CLI telemetry options"
❯ sqlite3 -batch ~/.config/tanzu-cli-telemetry/cli_metrics.db "select count(*) from tanzu_cli_operations;"
count(*)
3
❯ ./bin/tanzu context list --current
  NAME                ISACTIVE  TYPE   PROJECT  SPACE
  prem-ucp-intg-ctx7  true      tanzu

[i] Use '--wide' flag to view additional columns.
❯ sqlite3 -batch ~/.config/tanzu-cli-telemetry/cli_metrics.db "select count(*) from tanzu_cli_operations;"
count(*)
0

Some latency numbers with 6000 rows of metric data existing in local DB and if user opt-out, the metric data is deleted. Also captured the time taken when there are 0 rows in the DB. The latency numbers are in acceptable range.

❯ cp /tmp/test.db $HOME/.config/tanzu-cli-telemetry/cli_metrics.db
❯ sqlite3 -batch ~/.config/tanzu-cli-telemetry/cli_metrics.db "select count(*) from tanzu_cli_operations;"
count(*)
6000

❯ time ./bin/tanzu context list --current
  NAME                ISACTIVE  TYPE   PROJECT  SPACE
  prem-ucp-intg-ctx7  true      tanzu

[i] Use '--wide' flag to view additional columns.
./bin/tanzu context list --current  0.34s user 0.08s system 71% cpu 0.582 total

❯ sqlite3 -batch ~/.config/tanzu-cli-telemetry/cli_metrics.db "select count(*) from tanzu_cli_operations;"
count(*)
0
❯ time ./bin/tanzu context list --current
  NAME                ISACTIVE  TYPE   PROJECT  SPACE
  prem-ucp-intg-ctx7  true      tanzu

[i] Use '--wide' flag to view additional columns.
./bin/tanzu context list --current  0.33s user 0.08s system 70% cpu 0.586 total

Release note

Clear all the local metrics data when user opt out of CEIP

Additional information

Special notes for your reviewer

[vuil]

lgtm. I left a small suggestion but it is not blocking