ExprStringNameSpace replace / replace_all literal flag ignored for dataframes with multiple rows

[jameshar425]

Checks

• I have checked that this issue has not already been reported.
• I have confirmed this bug exists on the latest version of Polars.

Reproducible example

import polars as pl 

df=pl.DataFrame({'text':["I found <amt> yesterday."], 'amt': ["$1"]}) 

df = df.with_columns(pl.col("text").str.replace_all(f"<amt>", pl.col('amt'), literal=True).alias("text2"))
print(df)
# ┌──────────────────────────┬─────┬───────────────────────┐
# │ text                     ┆ amt ┆ text2                 │
# │ ---                      ┆ --- ┆ ---                   │
# │ str                      ┆ str ┆ str                   │
# ╞══════════════════════════╪═════╪═══════════════════════╡
# │ I found <amt> yesterday. ┆ $1  ┆ I found $1 yesterday. │
# └──────────────────────────┴─────┴───────────────────────┘

df2=pl.DataFrame({'text':["I found <amt> yesterday.", "I lost <amt> yesterday."], 'amt': ["$1","$2"]}) 

df2 = df2.with_columns(pl.col("text").str.replace_all(f"<amt>", pl.col('amt'), literal=True).alias("text2"))
print(df2)
# ┌──────────────────────────┬─────┬─────────────────────┐
# │ text                     ┆ amt ┆ text2               │
# │ ---                      ┆ --- ┆ ---                 │
# │ str                      ┆ str ┆ str                 │
# ╞══════════════════════════╪═════╪═════════════════════╡
# │ I found <amt> yesterday. ┆ $1  ┆ I found  yesterday. │
# │ I lost <amt> yesterday.  ┆ $2  ┆ I lost  yesterday.  │
# └──────────────────────────┴─────┴─────────────────────┘

Log output

$ POLARS_VERBOSE=1 python tmp2.py 
shape: (1, 3)
┌──────────────────────────┬─────┬───────────────────────┐
│ text                     ┆ amt ┆ text2                 │
│ ---                      ┆ --- ┆ ---                   │
│ str                      ┆ str ┆ str                   │
╞══════════════════════════╪═════╪═══════════════════════╡
│ I found <amt> yesterday. ┆ $1  ┆ I found $1 yesterday. │
└──────────────────────────┴─────┴───────────────────────┘
shape: (2, 3)
┌──────────────────────────┬─────┬─────────────────────┐
│ text                     ┆ amt ┆ text2               │
│ ---                      ┆ --- ┆ ---                 │
│ str                      ┆ str ┆ str                 │
╞══════════════════════════╪═════╪═════════════════════╡
│ I found <amt> yesterday. ┆ $1  ┆ I found  yesterday. │
│ I lost <amt> yesterday.  ┆ $2  ┆ I lost  yesterday.  │
└──────────────────────────┴─────┴─────────────────────┘

Issue description

Increasing the number of rows from 1-2 causes the bug to appear on both ExprStringNameSpace.replace and ExprStringNameSpace.replace_all.

I believe that this bug is only present when the "$" is in expr passed to the val parameter.

Expected behavior

The 1 row example performs how I would expect.

Installed versions

--------Version info---------
Polars:               1.5.0
Index type:           UInt32
Platform:             Linux-5.15.0-91-generic-x86_64-with-glibc2.35
Python:               3.10.10 (main, Jun 12 2024, 09:40:01) [GCC 11.4.0]

----Optional dependencies----
adbc_driver_manager:  <not installed>
cloudpickle:          3.0.0
connectorx:           <not installed>
deltalake:            <not installed>
fastexcel:            <not installed>
fsspec:               2024.6.1
gevent:               <not installed>
great_tables:         <not installed>
hvplot:               <not installed>
matplotlib:           <not installed>
nest_asyncio:         1.6.0
numpy:                1.26.4
openpyxl:             <not installed>
pandas:               1.5.3
pyarrow:              17.0.0
pydantic:             2.8.2
pyiceberg:            <not installed>
sqlalchemy:           <not installed>
torch:                2.4.0+cu121
xlsx2csv:             <not installed>
xlsxwriter:           <not installed>```

</details>

[corwinjoy]

I'm going to look into this over the next couple days to see if I can fix this.

[corwinjoy]

PR for fix. #19366

[corwinjoy]

And a followup. When I read this, I missed that the issue affects replace as well. This followup PR fixes the issue for replace as well.
#19413