Improve handling of renaming fable columns

Currently it drops the fable class, even if the renamed columns aren't special in anyway (key, index, distribution). It should be possible to rename all columns without issue, maintaining the fable class.

reprex provided by Eric Berger:

library(fpp3)
#> ── Attaching packages ──────────────────────────────────────────── fpp3 0.4.0 ──
#> ✓ tibble      3.1.6          ✓ tsibble     1.1.1     
#> ✓ dplyr       1.0.7          ✓ tsibbledata 0.4.0.9000
#> ✓ tidyr       1.1.4          ✓ feasts      0.2.2.9000
#> ✓ lubridate   1.8.0          ✓ fable       0.3.1.9000
#> ✓ ggplot2     3.3.5.9000
#> ── Conflicts ───────────────────────────────────────────────── fpp3_conflicts ──
#> x lubridate::date()    masks base::date()
#> x dplyr::filter()      masks stats::filter()
#> x tsibble::intersect() masks base::intersect()
#> x tsibble::interval()  masks lubridate::interval()
#> x dplyr::lag()         masks stats::lag()
#> x tsibble::setdiff()   masks base::setdiff()
#> x tsibble::union()     masks base::union()
N <- 20
a <- tsibble(date=2000+(1:N),x=runif(N),index=date)
a.lm <- a %>% model(tslm=TSLM(x ~ trend()))
a.fc <- a.lm %>% forecast(h=1)
hilo(a.fc,level=95)  ## outputs a tsibble
#> # A tsibble: 1 x 5 [1Y]
#> # Key:       .model [1]
#>   .model  date             x .mean                    `95%`
#>   <chr>  <dbl>        <dist> <dbl>                   <hilo>
#> 1 tslm    2021 N(0.45, 0.14) 0.453 [-0.2847969, 1.190116]95
### 
print(a.fc) ## a fable
#> # A fable: 1 x 4 [1Y]
#> # Key:     .model [1]
#>   .model  date             x .mean
#>   <chr>  <dbl>        <dist> <dbl>
#> 1 tslm    2021 N(0.45, 0.14) 0.453
class(a.fc) ## "fbl_ts" etc
#> [1] "fbl_ts"     "tbl_ts"     "tbl_df"     "tbl"        "data.frame"
colnames(a.fc)  ## ".model" "date" "x" ".mean"
#> [1] ".model" "date"   "x"      ".mean"
a.fc %>% rename(forecast = .mean)
#> # A tsibble: 1 x 4 [1Y]
#> # Key:       .model [1]
#>   .model  date             x forecast
#>   <chr>  <dbl>        <dist>    <dbl>
#> 1 tslm    2021 N(0.45, 0.14)    0.453
colnames(a.fc)[4] <- "forecast"
###
hilo(a.fc,level=95) ### Error
#> Error in `hilo()`:
#> ! Objects of type `tbl_ts` are not supported by `hilo()`, you can create a custom `hilo` with `new_hilo()`
#> • Objects of type `tbl_df` are not supported by `hilo()`, you can create a custom `hilo` with `new_hilo()`
#> • Objects of type `tbl` are not supported by `hilo()`, you can create a custom `hilo` with `new_hilo()`
#> • Objects of type `data.frame` are not supported by `hilo()`, you can create a custom `hilo` with `new_hilo()`
###
class(a.fc)
#> [1] "tbl_ts"     "tbl_df"     "tbl"        "data.frame"

^{Created on 2022-02-25 by the reprex package (v2.0.0)}

Start by running the supplied reprex, focusing on dplyr's rename operation and the hilo() entry point for the resulting forecast object. Done means renaming ordinary or special columns preserves the fable class and hilo(a.fc, level=95) continues to work afterward.